NINA KUNG v. WANG DIN SHIN

Judgment Summary (English Version)

Judgment Summary (Chinese Version)

FACV No. 12 of 2004

IN THE COURT OF FINAL APPEAL OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

FINAL APPEAL NO. 12 OF 2004 (Civil)

(ON APPEAL FROM CACV No. 460 of 2002)

_______________________

Between:

_______________________

Court: Mr Justice Chan PJ, Mr Justice Ribeiro PJ, Mr Justice Litton NPJ, Sir Noel Power NPJ and Lord Scott of Foscote NPJ

Dates of Hearing: 11 - 15 and 18 - 22 July 2005

Date of Judgment: 16 September 2005

_______________________

 J U D G M E N T

_______________________

Mr Justice Chan PJ:

1.  There is only one issue in this case, although by no means a simple issue.  It is this: Was the document which was written in Chinese, dated 12 March 1990 and purported to have been signed by Mr Wang Teh Huei (“Mr Wang”) and witnessed by a Mr Tse Ping Yim and another Mr Wang’s last will and testament? 

2.  Most regrettably, the trial went out of control and, quite unjustifiably, it took 172 days.  The trial judge, Yam J, held that this document together with three other documents were forgeries.  The majority of the Court of Appeal (Yeung and Yuen JJA, with Waung J dissenting) agreed with him.  Hence this appeal.

3. I have had the advantage of reading the judgment of Mr Justice Ribeiro PJ in draft.  I agree with his analysis, the reasons he has given and his conclusions.  I would just say a few words on the handwriting evidence. 

4.  In this judgment, I shall refer to the appellant, Mrs Wang as “the Wife”, the respondent, Mr Wang senior as “the Father”, the documents which were produced by the Wife and which included the document I have mentioned above as “the 1990 documents”, the signatures purporting to be those of Mr Wang appearing on the 1990 documents as “the Wang signatures”, those of Mr Tse Ping Yim (“Mr Tse”) as “the Tse signatures” and these signatures collectively as “the questioned signatures”.

Findings on handwriting by the lower courts

5.  The trial judge held that both the Wang signatures and the Tse signatures on the 1990 documents were forgeries.  In the Court of Appeal, Yeung JA upheld both findings.  Yuen JA agreed with the finding that the Wang signatures were not genuine, but was not satisfied to the required standard of proof that the Tse signatures had been proved to have been forged.  Waung J disagreed with the trial judge and held that both the Wang signatures as well as the Tse signatures were genuine. 

6.  At the hearing before this Court, the Father accepts the decision of the majority of the Court of Appeal (Yuen JA and Waung J) which set aside the trial judge’s finding that the Tse signatures were forgeries.  The present appeal has been conducted on the basis that the Tse signatures on the 1990 documents were genuine.

Effect of Court of Appeal’s finding re the Tse signatures 

7.  One of the Wife’s complaints is that Yuen JA failed to consider what effect her finding that the Tse signatures were not forged might have on the genuineness or otherwise of the Wang signatures.  It is submitted that she was wrong to speculate that there was a subtle form of forgery and to imply that the Tse signatures might have been placed on the 1990 documents afterwards, notwithstanding that there was no evidence to support this suggestion and that this was never the Father’s case.  It is contended that she should also have found that the Wang signatures were not forged.

8.  The Father’s experts expressed very strong opinions on the Tse signatures.  They were almost certain that the Tse signatures were forgeries.  Mr Gus Lesnevich had “no doubt” that they were.  Mr David Tsui Chee Keung said in his report that they “were not written” by Mr Tse.  He confirmed this in his evidence in court, saying that he would give them a -10 on a numerical scale of +10 to –10, which means that he was quite positive they were forgeries.  The Government handwriting expert, Mr Cheng Yau Sang, said in his report that the Tse signatures “were not written” by Mr Tse.  In his oral testimony, Mr Cheng said he was “very certain” of that conclusion.  Despite such strong opinions, Yuen JA and Waung J were not satisfied that the Tse signatures were forgeries.  And this is accepted by the Father.

9.  The same experts were however less confident about the Wang signatures.  Mr Lesnevich maintained that he had no doubt that they were forgeries.  In his report, Mr Tsui said that Mr Wang “did not write” these signatures, but in evidence, he was less certain and could only say that it was “likely” that they were forgeries, giving them a -5 on his scale between +10 and -10.  According to Mr Cheng’s criteria, there were four degrees of certainty: that a signature was (i) definitely not, (ii) highly probably not, (iii) probably not, or (iv) might not have been, written by the true author.  In his report, Mr Cheng took the view that the Wang signatures “might not have been written by Mr Wang”.  After he had had the opportunity of examining further samples of Mr Wang’s signatures, he was more convinced at the trial that the Wang signatures were “probably not” written by Mr Wang. 

10.  In my view, the effect of the Court of Appeal’s finding and the Father’s acceptance that the Tse signatures were not forged is two-fold.  First, the importance of the handwriting evidence relating to the Wang signatures is much reduced.  This is because Mr Tse had said in his statements that he saw Mr Wang sign the 1990 documents and then he signed as a witness.  As Lord Scott of Foscote said in his judgment (which I have read in draft and with which I agree), the focus is now shifted to the reliability of Mr Tse’s evidence. 

11.  The second effect is that the reliability of the experts’ opinion in respect of the Wang signatures has to be re-examined.  That the experts were held by the Court of Appeal and accepted by the parties to be wrong about the Tse signatures does not of course mean that they must also be wrong in their opinion about the Wang signatures.  After all, the Tse signatures and the Wang signatures were entirely different and had separately led the experts to draw their respective conclusions in respect of the two sets of signatures.  However, since the experts applied the same principles when examining the two sets of signatures and they were more certain about the Tse signatures than the Wang signatures, the reasons given by the experts for concluding that the Wang signatures were forgeries need to be reconsidered more carefully.

Handwriting evidence not the only evidence

12.  With regard to the handwriting evidence in this case, one important observation must be made.  The evidence on handwriting cannot and should not be considered as if there were no other evidence.  It is not the only evidence which goes to show whether the 1990 documents are genuine or not.  There is other evidence on this issue including in particular Mr Tse’s evidence in the form of two written statements.  The crucial question to be decided is whether Mr Tse’s evidence can safely be accepted in the light of all the evidence including the handwriting evidence.  If, having considered all the evidence, the court is satisfied to the requisite standard of proof that Mr Tse was telling the truth, then notwithstanding the handwriting evidence to the contrary, the court can still hold that the 1990 documents are genuine.  There are decided cases in which despite clear evidence from experts to the effect that the handwriting in question was forged, the court refused to accept it but favoured other pieces of direct evidence.

13.  In Newton v. Ricketts (1861) 9 HLC 263, which bears some resemblance to the present case, in exercise of a power under a marriage settlement, a deed of appointment was executed by a couple in the presence of two witnesses.  When the deed was subsequently disputed, one of the witnesses had passed away and the other swore an affidavit asserting positively to the due execution of the deed by the couple.  The signatures were challenged and an expert expressed the opinion that the signatures were forgeries.  However, the court preferred the evidence of the attesting witness who was not cross-examined and rejected the opinion of the expert. 

14.  In Fuller v. Strum [2001] WTLR 677, notwithstanding the opinion of a court-appointed single joint expert to the effect that the will under challenge had been forged, the court accepted the evidence of a live witness who testified that he saw the testator sign the will.

15.  It can be seen that the courts generally prefer the direct evidence of witnesses who actually saw and heard what happened to the opinion evidence of an expert.  This is especially so in the case of handwriting evidence because of the nature of such evidence.

Nature of handwriting expert evidence 

16.  A person’s signature can be recognised or identified by the writer himself or by someone who had witnessed the signing of the signature.  Mr Tse’s evidence falls within the second category.  This is proof by direct evidence.  In the absence of such evidence, this can be done by someone who is familiar with the signatures of the writer or by an expert in handwriting.  This is permissible under s. 17 of the Evidence Ordinance, Cap 8. 

17.  What an expert observes by examining the signature in question and comparing it with known and accepted signatures of the writer are matters of fact.  On the other hand, the drawing of a conclusion based on the facts which he has observed is a matter of opinion.  His opinion is admissible as evidence, subject to one qualification.  That is, his opinion need not be accepted by the judge or jury if they are not convinced by the reasons in support of such opinion.  The judge or jury can come to their own conclusions, based on the facts and their own reasoning. 

18.  However, an expert has 2 advantages over the judge and jury.  First, he has the scientific techniques or means to find out facts which a judge or jury may not be equipped to do.  Secondly, an expert has the necessary experience and expertise.  This can help him look for the relevant facts and come to a reasoned conclusion.  See Fuller v. Strum [2001] WTLR 677, 695 and Phipson on Evidence, para. 37–10 and 37-73.

19.  However, at the end of the day, it is for the judge or jury to form their own independent judgment with the assistance of the expert.  After all, they are the tribunals of fact.  The duty of expert witnesses is, as Lord President Cooper said in Davie v. Edinburgh Magistrates [1953] SC 34 at 40: “to furnish the Judge with the necessary scientific criteria for testing the accuracy of their conclusions, so as to enable the Judge or jury to form their own independent judgment by the application of these criteria to the facts proved in evidence.”

20.  Given the nature of expert evidence, one has to approach such evidence with caution.  While the factual part of the evidence of an expert is verifiable and is reliable or unreliable as any other piece of factual evidence, the opinion part of an expert’s evidence is totally different.  Handwriting analysis is not an exact science and the opinion of a handwriting expert, however objective it is, is inherently less precise than a conclusion based on the results of a scientific analysis.  The acceptability of an expert’s opinion depends very much on how sound and convincing the reasons for his opinion are. 

The Wife’s submissions on handwriting 

21.  The Wife submits that the trial judge and the majority of the Court of Appeal (Yeung and Yuen JJA) were wrong in holding that the Wang signatures were not genuine.  Her main submission on the handwriting evidence is that there was cogent expert evidence showing that the Wang signatures were genuine: (a) the Wang signatures were written with naturalness; (b) there were significant similarities between the Wang signatures and the sample signatures; and (c) there were matching variations between the Wang signatures and the sample signatures.  It is contended that the trial judge and the majority of the Court of Appeal failed to apply objective criteria and were wrong to rely on their own subjective impression in finding that there were significant differences.

Principles of examination and comparison

22.  The principles which should be applied in determining whether the questioned signatures were genuine or forged were fully discussed at the trial.  These principles have been set out in detail in the judgments in the courts below.  A comprehensive summary was given by Waung J in paragraph 1163 of the Court of Appeal judgment.  None of these principles are seriously disputed by the parties.  It is only their application to the signatures in this case and the conclusion reached which are the bone of contention.  For the present purpose, I shall only state the most crucial points which can be gathered from the expert literature.

23.  Each person has his own writing habit.  Because of such habit, there are bound to be similarities among his signatures.  These similarities are the features of his signatures.  They are individual characteristics which are only personal to him.  On the other hand, no two signatures written by the same person can be exactly the same.  There are bound to be differences.  The number and quality of differences depend on the conditions of the writer and the surrounding conditions.  But since the signatures come from the same person, these differences would be within a range and would be regarded as his normal variations.  These normal variations also form part of his individual characteristics.

24.  In order to decide whether a questioned signature is genuine or not, it is important first to identify the individual characteristics of the writer which represent his writing habit from samples of his genuine signatures.  This may not be easy if there are not sufficient available samples.  Individual characteristics which are inconspicuous “should be given the most weight; for these are likely to be so unconscious that they would not intentionally be omitted when the attempt is made to disguise and would not be successfully copied from the writing of another when simulation is attempted.” (See Osborn’s Questioned Documents, p. 250.)

25.  These individual characteristics should then be compared with the questioned signature to see whether there are any similarities and differences between the questioned signature and the samples of genuine signatures. 

26.  If there are significant differences or “divergences in amount and quality beyond the range of variation … that cannot reasonably be accounted for by changed conditions in the writer or surrounding the writer”, one can draw the conclusion that a signature is not genuine.  Conversely, the absence of fundamental differences (as opposed to trivial differences) together with the presence of a combination of a sufficient number of similarities with individual qualities and characteristics can form the basis of a conclusion that the signature is genuine.  “The process is always a double operation, positive and negative, and if error is to be avoided neither part of the process should be overlooked.”  See Osborn, p.205-251. See also “Scientific Examination of Questioned Documents” by Ordway Hilton, p.174.  In the comparison exercise, it is important to bear in mind that “mere similarities do not necessarily prove genuineness any more than mere superficial differences necessarily prove lack of genuineness.” See Osborn, p.241.

27.  At the end of the day, it is necessary to make a “consolidated evaluation” of all the evidence in order to come to a conclusion whether it can be said that a particular signature is genuine or forged.  

Knowledge of the Chinese language

28.  Documents A, B and C and the questioned signatures on the 1990 documents were written in Chinese.  It has been suggested that since Mr Lesnevich was admittedly not conversant with the Chinese language, his ability to give an expert opinion on the genuineness or otherwise of the questioned signatures is limited.

29.  In an article “A comparative approach to the examination of Chinese handwriting. Part 5 – qualitative parameters” published jointly by S C Leung, W L Cheung, H T Fung and Y L Cheung in the Journal of the Forensic Science Society in 1993, the special features of Chinese characters are highlighted.  The authors take the view that:

“The same features are applicable to English handwriting, in spite of the differences in structure and writing movements associated with the two languages.”

30.  F Brewester, in his book “Contested Documents and Forgeries” considered that knowledge of a language other than English is not necessary when dealing with signatures in such other language, but an intensive study of the manner in which the characters are made would be desirable.  See p.105.  

31.  In Kung Wong Sau Hin v. Kung Kwok Sun and others, Probate Action No. 2 of 1982, an English handwriting expert was called to give evidence regarding a Chinese will.  Deputy Judge Nazareth QC, having heard the experts and submissions on the matter, came to the view that:

“… it is possible for an occidental questioned document examiner to make meaningful examinations of documents written with ideographic characters.”

32.  As a general proposition, I would accept that the principles of examination and comparison are similar and the psychological factors underlying a forgery are the same for all languages.  However, since different languages have different characteristics, when applying these principles to Chinese handwriting, particular attention must be paid to these characteristics.  Knowledge of the language in which the alleged forgery is written is a clear advantage.  It would not be easy for someone who does not know that language to look for the relevant characteristics in the handwriting in order to perform an effective examination and comparison.  Hence, although Mr Lesnevich had much experience and expertise in English handwriting, he was handicapped when it came to Chinese handwriting.

Features in Chinese handwriting

33.  Each Chinese character has a unique structure, consisting of one or more radicals (部首), each radical having a particular position in relation to the other radicals and to the whole character.  Further, there is a sequence of each of the strokes in a character which people learning how to read and write Chinese are taught to follow.  Very often, the structure of a character is written in a simplified way, certain strokes are conveniently omitted and the orthodox sequence is not followed.  This is particularly the case in signatures since there are different personal reasons why a person does not want to follow the structure of the characters or the orthodox sequence of the strokes when it comes to writing his own signatures.  The way a person writes a character, whether following thestructure of the character and the orthodox sequence of the strokes or adopting his own structure or sequence, constitutes his personal style and becomes his own writing habit.

34.  The structure and style in which a Chinese character is written, particularly those which are unique or personal to the writer, are important features relevant in the examination and comparison exercise in determining whether a particular piece of handwriting or signature is genuine or forged.

35.  Another important feature involves the pen movements of the handwriting or signature.  “How the pen movements are executed together with a detailed study of the relative positions of the strokes reveal the writing habits of the writer and are basic to the examination of the handwriting.”  See the joint article: “A comparative approach to the examination of Chinese handwriting. – the Chinese character” at p. 261.  Since Chinese characters generally have a more complicated structure, the pen movements adopted in writing Chinese characters are usually slightly more sophisticated than that used in writing English.  The presence of similar pen movements particularly in relation to features which are unique or personal to a writer would be a very strong indication of genuineness.

36.  Thequality of the writing and the pictorial appearance of the characters are also important attributes.  So are the writing speed and style.  See p.18 of the joint article “A comparative approach to the examination of Chinese handwriting, Part 5 – questioned parameters”:

“In Chinese handwriting, writing speed and style have a significant effect, not only on the pictorial appearance but also on certain personal attributes.  … Qualitative parameters can also be affected by writing speed which in the Chinese script would have an effect of causing the strokes to become merged.  As a result, the writing sequence of some strokes can change.  At a high writing speed, the stroke also tends to be more rounded, resulting in features different from those observed in the handwriting produced at a different speed.” 

37.  In applying the principles of examination and comparison applicable to English handwriting, the differences in features which are special to Chinese handwriting and signatures must also be borne in mind.

Sample signatures used for comparison

38.  The questioned signatures were compared by the experts with 18 samples of Mr Wang’s signatures.  An analysis of the 18 sample signatures reveals the following data which are relevant for present purposes.

39.  First, of these sample signatures, two were executed in 1958, one each in 1967, 1975 and 1980, two in 1983, four in 1984 of which two were executed on the same occasion, and seven in 1985 of which three were executed on the same occasion and four were executed on another occasion.  The fact that a diligent search could only find not more than 18 sample signatures (executed on 12 occasions) for the period between 1958 and 1985, would suggest that Mr Wang did not have too many opportunities to sign in Chinese over the years.  Some of the experts at one stage or another were concerned that there were not sufficient samples for comparison purposes and this might have affected the reliability of the results of their comparison. 

40.  Second, three of the sample signatures were written by fountain pen and they included the 1958 signatures; five by ball pen or ball point ink; and the remaining 10 by aqueous ink (which could be fountain pen ink or roller ball ink).  The Wang signatures which bear close resemblance to the 1958 signatures were written by an instrument such as a calligraphic pen (with a broad nib) similar to a fountain pen which was the type used to write the 1958 signatures. 

41.  Third, 16 of the sample signatures were originals and two were photocopies.  All the samples produced by the Wife (11 of them) were originals and the more recent signatures which were executed in 1984 and 1985. The original samples containing the two 1958 signatures were among the samples kept in the possession of the Father, not the Wife. 

The Wang signatures generally

42.  When the Wang signatures are compared with the sample signatures, the first impression one gets is that they resemble the 1958 signatures more than the more recent signatures.  The trial judge remarked that if the 1958 signatures were discarded, the difference between the Wang signatures and the other sample signatures is such that “a fair minded person will have to conclude that they are different writings”.

43.  But with handwriting, while subjective impression is usually the starting point, objective criteria are the ultimate guide.  It is necessary to examine both the Wang signatures and the sample signatures more closely. 

Orientation and alignment 

44.  In his macro analysis of the handwriting evidence, the judge considered that the Wang signatures were different from the post-1967 signatures in their orientation and alignment.  He took the view that this was an indication that the Wang signatures were not genuine.  Yuen JA was also greatly concerned by this feature.  It is clear that she considered this as a significant difference and a weighty factor when drawing her conclusion that the Wang signatures were forgeries.

45.  The orientation and alignment of the Wang signatures was not raised as an issue until Professor Jia, the Wife’s expert on handwriting, was cross-examined.  To illustrate the difference in orientation and alignment, a method was used by the Father’s experts and that is to draw a rectangular box marking the external bounds of the whole signature with a vertical line down the middle of the box.  Using this method, it can be seen that most of the Wang signatures are on the left side of the central vertical line, whereas all sample signatures except the 1958 signatures (and the 5 signatures which were written horizontally and cannot be used for comparison for this purpose) are more or less in the middle of the central vertical line.  For the 1958 signatures, at least sample 669, they are also on the left side of the central vertical line but not as much as the Wang signatures.

46.  However, Professor Jia denied that there was such a difference.  He considered that the Father’s method of measuring the alignment was not appropriate to illustrate the orientation of the signatures since the lines of the box so drawn might be affected by the style and speed of the writing and the change in the extremities of the characters.  The method he adopted is done by drawing a vertical line along the vertical stroke of the 1^st character “Wang”（王） in each of the signatures down the whole signature, and then drawing another line joining the furthest right parts of the 2^nd and 3^rd characters “Teh” （德）and “Huei”（輝）.  Using this method, Professor Jia sought to demonstrate that the two lines so drawn show the actual orientation of the whole signature and that such orientation is similar in all the signatures.  The trial judge rejected this method, describing it as arbitrary. 

47.  The purpose of looking for the alignment of the characters in a signature is to examine the orientation of the main structure of the whole signature.  This depends on the relative positions of the different radicals in each of the characters in the signature as written by the author.  It can be influenced by the style and speed of the handwriting (as Professor Jia said) and by the space available, the paper on which the signature is put and the position of the writer (as suggested by Mr Lesnevich).  Professor Jia’s method was simply a genuine attempt to achieve this purpose and was not without any basis.  It was not any more arbitrary than the method adopted by the Father.

48.  But the point is not which of the two methods can more accurately demonstrate the orientation of the signatures.  The point is that this was never properly canvassed by the experts, let alone relied on as an indication of forgery.  In any event, neither Mr Lesnevich nor Mr Cheng considered that the difference in orientation and alignment in this case was of great significance and Mr Tsui agreed that the whole structure of the Wang signatures was comparable to the sample signatures.

49.  Even if the judge could properly reject Professor Jia’s evidence on this matter, it was wrong for him to have held in the absence of expert evidence that there was a significant difference in orientation and alignment.  He was not entitled to rely on this alleged difference as an indication of forgery.  Unfortunately, Yuen JA fell into the same error.

General appearance and outlook 

50.  The Wang signatures on the 1990 documents are clearly similar to the 1958 signatures but different from the other sample signatures.  If the Wang signatures were genuine, it does seem strange that Mr Wang would have signed in 1990 in a way which resembles what he did 32 years ago.  It must be noted that in 1958, Mr Wang was a young man of 24.  He was then at the beginning of his business career.  His signatures appeared to be more sophisticated and, as Mr Lesnevich described, more artistic.  A comparison between the 1958 signatures and his more recent signatures reveals that throughout the years his signatures had become more simplified and less sophisticated both in terms of structure and style. 

51.  Mr Lesnevich’s conclusion that the Wang signatures were forgeries was based on the premise that Mr Wang had a better controlled and more artistically written signature in 1958 and that since 1967, his ability to do so had degenerated to the extent that he had lost his ability to sign his signature in Chinese with the same high degree of writing skill and ability as that in 1958.  However, none of the other experts was willing to go as far as to say that Mr Wang had lost his ability to sign as he did in 1958, although they agreed that his signatures had clearly changed.

52.  As Wilson R Harrison said in “Suspect Documents, Their Scientific Examination” at p. 388:

“Considerable doubt can be cast upon the authenticity of a suspect signature when it can be shown to be of a pattern which had been discarded by the writer before the date attributed to the document on which it appears, and even less credence will be given to a signature which considerably antedates the style of genuine signature to which it conforms.”

53.  Although there is no evidence that in 1990, Mr Wang had “discarded” his 1958 pattern of signature, the resemblance between the Wang signatures (which were supposed to have been executed in 1990) and the 1958 signatures is legitimately a matter of great concern and a weighty factor against genuineness.  If they were genuine, why would Mr Wang sign in such a style?

Explanations for resemblance to the 1958 signatures

54.  The Wife through her experts and counsel, offered a number of explanations for the unusual resemblance.  The trial judge rejected all of them.  Nor were Yeung and Yuen JJA satisfied with them.  Yam J speculated that since the originals or copies of the documents containing the 1958 signatures were kept by the Wife, she had the opportunity to make use of them for simulation.  Yeung JA even suggested that the simulation could have been made from signatures in other documents written in the 1950s.  But as I have mentioned earlier, the originals of the documents containing the 1958 signatures were in the Father’s possession: they were produced by Mr Yih Lee Kong, a witness called by the Father.  It is not clear when the Wife came to have copies of the documents containing the 1958 signatures.  But the real problem lies with the approach adopted by Yam J and Yeung JA.  There is no evidential basis for the suggestions made by them.  In making their respective suggestions, they seemed to have assumed that the Wang signatures were forged and were speculating how the forger had managed to do it.  

55.  One of the explanations by Professor Jia was that Mr Wang had two different ways in writing his signatures: one represented by the 1958 signatures and another represented by those post 1967 signatures.  This was rejected by the trial judge on the basis that a person would not normally maintain two different styles separated by a span of 30 years.  This might be right, although the evidence regarding Mr Wang’s style was scanty because there were so few samples of his Chinese signatures.  Furthermore, the chart of signatures clearly shows a gradual change (or evolution, as Yeung and Yuen JJA described it) in Mr Wang’s signatures. 

56.  Professor Jia then suggested that the Wang signatures were written in a “running style with some clerical script characteristics”.  I do not think this is convincing either because this still does not explain why Mr Wang would choose to adopt such a style after so many years.  It must also be borne in mind that the small number of available sample signatures suggests that he did not have too many occasions to sign in Chinese and yet all of the sample signatures, except the 1958 ones, are not in this running style with clerical script characteristics. 

57.  Counsel for the Wife also made several points.  First, it is said that there are features in the Wang signatures which can be found in the more recent sample signatures but not in the 1958 signatures.  Secondly, it is argued that there were variations in the 1958 signatures which are scattered amongst the questioned signatures.  Thirdly, resemblance with the 1958 signatures can be explained by the differences in style and speed in executing the questioned signatures.  Fourthly, the questioned signatures were written by a writing instrument which is similar to the one used to write the 1958 signatures.

58.  It is sought to demonstrate the first point by five examples which show features that are present in other samples although not in the 1958 samples.  Presumably, counsel was seeking to show that the Wang signatures do not merely resemble the 1958 signatures.

59.  I do not think these examples can illustrate the first point counsel is making.  First, each of the five examples refers to a particular feature in only one of the four questioned signatures.  While the feature taken in each example is not present in the 1958 samples but in some of the post 1967 samples, the other three questioned signatures do exhibit features appearing in the 1958 samples but not in the other samples.  I should think that these examples are more appropriate to illustrate the matching natural variation point which I shall deal with later in this judgment.  Secondly, the features in three of the examples (examples (1), (2) and (4), using the Wife’s numbering) are readily observable, can be easily imitated and hence do not carry much weight.  Another example (example (5)) is not well defined.  Only one example (example (3)) is unique and inconspicuous, but as I said, does not illustrate the point made by counsel.

60.  In respect of the second point, the fact that variations found in the 1958 signatures are scattered among the questioned signatures only reinforces the fact that the questioned signatures resemble the 1958 signatures.  Again it does not explain why this is so.

61.  The third point raised by counsel is, in my view, more convincing.  The nature of the paper used can affect the writing speed and the writing speed does have an effect on the style of the writing.  In the joint report compiled by the Wife’s experts, it is said that: “in Chinese handwriting, writing speed and style have a significant effect, not only on the pictorial appearance but also on certain personal attributes”.  Mr Tsui agreed with this comment.  And there is the evidence that the Wang signatures were indeed written more slowly.  The 1990 documents were written on flimsy paper.  The Wang signatures were more structured and artistic than the rather simplified structure and casual style of the more recent signatures.  These matters were apparently not fully appreciated by the trial judge as relevant to the point in issue. 

Use of the calligraphic pen 

62.  The fourth point raised by counsel is that instruments (fountain pens) used to execute the 1958 signatures were similar to the type of calligraphic pen which is said to have been used to sign the Wang signatures and this is offered as a reason to account for the resemblance between the two sets of signatures.

63.  None of the experts had directly dealt with this point although they had commented on the effect of the use of a calligraphic pen on individual dissimilarities.  Mr Lesnevich was of the opinion that the dissimilarities were not the result of the writing instrument.  But then, he did not read or write Chinese and his opinion on this aspect has to be considered with caution.  Mr Tsui accepted that the calligraphic pen could have accounted for the upward movement of the 1^st horizontal stroke in the character “Wang” （王）although not the degree of tilting.  Apart from that, he did not make any further comment on the use of such writing instrument.  Mr Cheng mentioned that one of the reasons for the qualified opinion given in his report was the difference in the writing instruments.  However he did not elaborate on it.  

64.  Yam J did not deal with this point although counsel for the Father argued that by implication, he must have rejected it.  In the Court of Appeal, Yeung JA did not accept that the use of such a pen could have resulted in the significant differences between the Wang signatures and the post 1967 signatures.  However, the examples he gave in support of this conclusion did not illustrate the point but only dealt with why he did not consider such writing instrument could have caused certain features in the Wang signatures.  Yuen JA commented on the use of a calligraphic pen only in relation to a particular dissimilarity.  Waung J dealt generally with the implications of the use of a calligraphic pen in the context of forgery.  I find his comments, particularly those in paragraph 1200, very helpful.

65.  It is accepted that the type of writing instrument used would have an important effect on the handwriting.  See Osborn, p.386.  The effect “extends beyond that of mere difference in appearance”.  See Harrison, p. 331.

66.  The use of a calligraphic pen with a broad nib will reduce the writing speed and this is borne out by the evidence that the Wang signatures were indeed written more slowly.  The use of such instrument will also affect the writing style as the writer will be more inclined to write more artistically instead of more cursorily.  And it is clear that the 1958 signatures were written more artistically and the post 1967 signatures more cursorily.  Counsel for the Wife gave three examples to show the effect of the use of calligraphic pen on the appearance and style of the writing: (1) the first horizontal stroke and (2) the shape of the loop in the middle of the character “Wang” （王）, and (3) the shape of the hook in the left radical “光”of the character “Huei” （輝）.  I am inclined to agree.  These examples do illustrate the effect of the calligraphic pen.

67.  In my view, the use of a calligraphic pen could be a reasonable and valid explanation for the resemblance.  But this was unfortunately not canvassed at the trial.  The trial judge and the majority of the Court of Appeal had failed to consider whether the use of a calligraphic pen could have been a relevant matter in the present case and if so, what effect this would have. 

Naturalness

68.  Naturalness found in a signature is regarded as a very strong indication of genuineness.  The rationale is that a forger would normally concentrate on the form of the writing, trying his best to make his work look genuine whereas the writer of a genuine signature is simply doing it as part of his writing habit, giving allowance for the difference between doing ordinary handwriting and signing a signature.  Harrison explained this in greater detail at p. 399 of his book:

“… a genuine signature, particularly that of a practised writer, is written fluently and without conscious thought being given to the exact form of the letters, simply because of the familiarity of the writer with the task he is accomplishing.

In marked contrast, … when a form is produced from memory, or perhaps copied from one which is before the writer, to form a simulated forgery, there can be no swift, carefree movement of the pen.  Unlike the casual writer, the forger has to concentrate on the task in hand if he is to ensure that his pen point is reproducing the letter designs he has memorised or is observing with half his attention whilst giving the remainder to the writing.

Instead of a swift and sweeping movement of the pen, there has to be a slow and careful motion which is the product of care from the moment the pen is cautiously applied to the paper until it is finally lifted from its surface.  The forger has a double care; not only has he to copy the handwriting of another, but he must continually be on the alert to suppress his own handwriting instincts.  Therefore that which is finally committed to the paper, be it either traced or simulated forgery, cannot be termed ‘handwriting’ in a normal sense of the word; it is more accurately described as a careful and painstaking ‘drawing’ of the handwriting of another person.”

69.  It would seem that evidence of pauses or hesitations and changes of pen pressure, direction and speed would be signs of unnatural writing and hence indicative of forgery.  On the other hand, if the writing is natural and free flowing with what can be described as “flying starts” and “flying finishes” or intermediate strokes where the pen comes off the paper or connecting strokes which show continuity of pen movements, these would be regarded as indications of unconsciousness of details pointing to genuineness.  See Osborn 364.

70.  The joint report prepared by the Wife’s handwriting experts described the writing in the Wang signatures as follows:

“The structures of the signatures … are normal, the stroke movements are natural, the degree and pace of pen pressure are clearly rhythmic, and the paper was even scratched and cut through when there was heavy pressure.  By making use of the stereo microscope and infra red vidicon comparer…, no features of imitation have been found in the writing, such as pauses in or rewriting of strokes or rigid and awkward, or apparently fast but in fact slow stroke movement.”

71.  When this part of the joint report was referred to Mr Lesnevich, he agreed with these comments.  He added that by and large, there were no such features as pauses in or rewriting of strokes or rigid or awkward movements.  He also said:

“They were too well written; the skill level was very good; there was a fluidity to them.  That was a naturalness.

It was obvious after going through the comparisons, that they were naturally written, other than maybe a little hesitation in a few points, that they were fluid, that they were artistically well done.” (Transcript day 18, pp. 171 and 172)

72.  Mr Tsui considered that the Wang signatures were not written too slowly, but because they were not written fast enough, there was some unevenness in a few places.  However, he added that the writing in the Wang signatures was not of an inferior quality, but admitted that this aspect was not too important and he would still regard the Wang signatures as not too bad.  Mr Cheng accepted that there was no retouching in those signatures.  Under cross examination, Professor Jia also confirmed that they were written naturally. 

73.  So it is clear, and there is support from the experts (except one guarded qualification made by Mr Tse), that the writing in the Wang signatures does exhibit naturalness and a free flowing style.  Such naturalness and style of writing is inconsistent with that of a person trying to commit a forgery.  For the reasons suggested by the academic writers, these features strongly support the conclusion that these signatures are unlikely to be forgeries.

74.  However, the trial judge did not consider that the Wang signatures were written naturally and he based this conclusion on (1) what he regarded as retouching and (2) the presence of tremors.  Was he justified to come to that conclusion?

Retouching in the writing - judge playing expert?

75.  The question of retouching as a sign of unnaturalness came up during the cross-examination of Professor Jia.  He was using a microscope to illustrate his opinion that there were special features in the questioned signatures which pointed to genuineness when it was put to him that there were “signs of retouching” in the Wang signatures. 

76.  In his judgment, the judge pointed out nine instances in the Wang signatures where he said he could notice retouching or unnatural writing.  It is to be noted that this part of his judgment was copied substantially from the written submissions of counsel for the Father.  For two reasons, I do not propose to set out these instances of retouching.  First, none of the experts called by the Father considered that there was any retouching or evidence of unnatural writing.  Mr Lesnevich said there was no “feature of imitation” such as retouching in the present case.  Mr Cheng confirmed that he could see no retouching.  Mr Tsui made no mention of any retouching in the Wang signatures but accepted that there are several places of ink diffusion in those signatures.  Secondly, the only evidence directly dealing with each of these instances of retouching came from Professor Jia.  All of them were denied by him.  In respect of the 8^th instance (according to the order as presented by counsel) it was said that this occurs in the character “Teh” （德） in Document C: the first slant in the left radical “彳”appears to be a large dot.  Professor Jia explained that this was not a retouching but was possibly done by more than one stroke such as the writer trying to refine it and this might be due to a number of reasons. 

77.  Counsel for the Wife submits that the judge was playing expert by applying his own observations and finding signs of retouching which the experts could not identify.  In defence of the trial judge, Yeung JA said the trial judge was entitled to rely on his own observations because of the inquisitorial nature of the probate action and could disagree with Professor Jia.  Yuen JA did not make any comment on this criticism. 

78.  Neither Yeung JA nor Yuen JJA who found that the Wang signatures were forgeries dealt with these instances of retouching.  They did not express any view as to whether they agreed or disagreed with the trial judge.  The only remark made by Yeung JA was on the first instance mentioned by the judge and that was about Professor Jia failing to draw the judge’s attention to the possibility of an optical illusion when the Wang signatures were placed and magnified under a microscope. 

79.  Nor did counsel for the Father make any substantial argument either in their written case or in oral submission on these instances of retouching, except suggesting that the 8^th instance which I have mentioned above was in effect admitted by the Professor Jia, when in fact he did not made such admission.

80.  In respect of the so-called retouching or unnatural writing, I would only make the following observations.  First, each of the characters in the questioned signatures is only about ½ inch in size.  When it was shown under the microscope, it was magnified many many times.  Although the parties are unable to tell us now how much it had been enlarged when the signatures were seen under the microscope, the photographs produced in court confirm this to be the case.  Needless to say, the features tend to be very much exaggerated.  Take the 8^th instance as an example.  The actual size of the slant which was said to have been retouched is only 1/32 inch, just a little more than a dot.  In the photograph shown on page 28 of the Power Point Bundle, that becomes a patch of 1½ inch by 1¼ inch!

81.  Secondly, one has to bear in mind that the writing instrument is a calligraphic pen, the paper is flimsy and the ink might have caused diffusion.  Not only Professor Jia but also Mr Tsui accepted that there was ink diffusion in some parts of the questioned signatures.

82.  Even if it can be said that the judge was entitled to apply his own observations, I would say that it cannot be said with any reasonable degree of confidence that there is any retouching or unnatural writing in these signatures.  With the kind of pen, paper and ink as used on the 1990 documents, and when the signatures were examined under such high degree of magnification as shown in the photographs contained in the Power Point Bundle, the pictorial appearances might have been distorted and it is not surprising to see under the microscope a closed loop in instances (1) and (5) as pointed out by the judge, a break in the stroke in instance (2); a bulge or patch or retouching in instances (3), (4), (6), (8) and (9).  As to the twisting in instance (7), this is so obvious that it looks more like a mistake made during the execution of the stroke.  It is common that a stroke or a dot is accidentally missed when a character particularly in a signature is written naturally and freely. 

83.  In my view, the fundamental problem is that in the absence of positive evidence from any of the experts, there is simply no evidential basis for the judge’s finding that there was retouching or unnatural writing in the Wang signatures.  Without such finding, there is nothing to counter the unchallenged evidence that the Wang signatures were written naturally and with a free flowing style which are significant features pointing towards genuineness.  It would seem that the whole exercise is an example of the judge having concluded that the signatures were forgeries was looking for material to support that conclusion, adopting counsel’s written submissions almost verbatim in the process.

Presence of tremors

84.  It is not disputed that there are minor tremors in the Wang signatures which do not appear in the sample signatures.  The presence of tremors may suggest that the writing is not natural and that unless there is an explanation for it, such as ill health or injuries, this may be regarded as an indication that the signatures are not genuine.

85.  However, none of the experts considered the minor tremors in the present case to be of great significance.  Professor Jia said that the tremors were not the result of forgery since the signatures were written freely and naturally.  He further explained that if these tremors were associated with forgery, they should be found at the more difficult strokes and not in the long horizontal or vertical strokes as in the present case. 

86.  Notwithstanding such opinions, the judge somehow considered these tremors as highly relevant to the issue of forgery.  In rebutting the explanation suggested in the joint report by the experts for the Wife that the tremors might be caused by Mr Wang’s injuries, the trial judge went through the medical evidence on Mr Wang’s injuries at great length.  He concluded in paragraph 11.18:

“On the other hand, the very minor nature of the tremors found in the strokes of the questioned signatures would tally with the kind of tremors caused by psychological cause, such as being anxious or exciting, which would probably be the sort of psychological condition only of either the forger or Wang himself.  Consequently the medical evidence cannot support the contention that these tremors were caused by Wang only.  Since I found for those reasons in the last chapter that the four questioned signatures are forged signatures of Wang, I would also find that probably the tremors were caused at least partly by the psychological condition of the forger rather than Wang himself.  They are also tremors of the forger when he tried to slowly simulate the signature of Wang according to his two models.”

87.  I must confess I do not understand the logic of this conclusion.  It is tantamount to saying that “because I have found these signatures to be forgeries, I now hold that these tremors are signs of forgery.” This approach is blatantly wrong.  The presence of tremors if they were of any significance, should have been considered in the trial judge’s “last chapter” (i.e. chapter 10 of his judgment) together with other evidence before he came to the conclusion that the Wang signatures had been forged.  His approach renders the whole of chapter 11 superfluous.  It follows that the judge’s remark that the tremors were signs of the forger cannot be supported. 

88.  It is to be noted from the passage in the judgment quoted above that the other alternative explanation for the presence of tremors which the trial judge had in mind is the psychological condition of Mr Wang, such as “being anxious or exciting”.  I am not surprised that that might at least be one of the possible reasons for the tremors.  First, Mr Wang had just had an accident two days before and was hospitalised.  He discharged himself against medical advice and went back to work in the office.  And it was in the evening after normal office hours.  Secondly, there is evidence (from Mr Tse in his second statement) to the effect that Mr Wang was rather moody on that evening. 

89.  The judge had obviously regarded the presence of tremors as indicating signs of forgery or at least negativing the evidence of naturalness in the Wang signatures.  This view is clearly unfounded.  For obvious reasons, none of the Court of Appeal judges dealt with this matter.  The result is that the evidence that the Wang signatures were executed naturally and in a free flowing style remains intact and is a strong indication of genuineness.

Significant similarities and dissimilarities

90.  All the experts agreed that it is important to identify the significant similarities and differences (or dissimilarities) between the Wang signatures and the known samples.  Admittedly there are some similarities and dissimilarities.  What the experts did not agree was (1) whether a particular feature is a similarity or dissimilarity, and (2) whether these similarities and dissimilarities were significant or not.  What one party regards as a significant similarity is considered by the other party as a significant difference.

91.  The experts accepted that a consolidated evaluation must be conducted, weighing all the similarities and dissimilarities together before coming to a conclusion.  Apparently the trial judge failed to do that.  For in his conclusion, he said (paragraph 31.37):

“All the similarities pointed out by the defendant’s experts were only superficial similarities produced by simulation.  There are many unexplained but significant differences.  There are also other features including tremors, retouching and retracing, and different writing movements with the sample signatures of Wang.  All these are features and indications of forgery. The evidence in support of such a finding of forgery is very strong and cogent.” (emphasis added)

92.  This passage demonstrates that the judge was wrong in two aspects.  First, it seems he had assumed forgery and rejected all the similarities on the ground that they were forged by simulation even before any consolidated evaluation.  Second, he said he had also taken tremors, retouching and retracing into consideration but as discussed above, there is no basis for doing that.  The judge’s conclusion is therefore flawed. 

93.  Counsel for the Wife provides some examples of significant similarities to support the genuineness of the Wang signatures.  To mention only those addressed to the Court in submission, they are: (1) the subtle pen wielding movement at the start of the vertical stroke in the character “Wang” （王）; (2) the first two strokes of the character “Wang” （王）; (3) the extra dot of the radical “心” in the character “Teh” （德）；and (4) the subtle pen wielding movement at the start of the first vertical stroke in the left radical “光”in the character “Huei” （輝）. 

94.  In my view, these examples illustrate some of the similarities with the sample signatures.  However, I do not agree that they are subtle or inconspicuous (in the case of (1), (2) and (4)) or well defined (in the case of (3)).  Such similarities do not seem to be significant.

95.  As to dissimilarities, the judge adopted a micro analysis.  A number of dissimilarities are pointed out to the judge by the experts either in their reports or in their oral evidence in court.  The judge adopted most of the submissions for the Father and held that they were significant dissimilarities which indicated that the Wang signatures were forgeries.  Yeung JA discussed some of examples and Yuen JA dealt with a few.  They too held that they were significant dissimilarities.  Waung J dealt with most of them in detail but came to the opposite conclusion.  The whole exercise is, however, not a game of numbers.  It is the quality of the dissimilarities that counts.

96.  In relation to the character “Wang” （王）, the dissimilarities relied on by the judge are these.  (1) The first horizontal stroke is longer and slanting upwards and not horizontal or slanting downwards. (2) There is a gap between the 1^st horizontal stroke and the beginning of the vertical stroke.  (3) The starting point of the vertical stroke is very much to the left of the mid point in the first horizontal stroke and not in the middle. (4) The length of the vertical stroke above the loop representing the second horizontal stroke is rather short.  (5) The vertical stroke is slightly slanting from the right to the left and not upright. (6) The loop in the middle formed by the vertical stroke and the second horizontal stroke is virtually closed.  (7) The second horizontal stroke is on the rise when it crosses the vertical stroke so that the highest point which resembles the head of the numeral “2” is on the right of the vertical stroke and not on the left or in the middle.  (8) The beginning of the third horizontal stroke is a continuation of the slanting downward slightly to the left movement of the second horizontal stroke with retracing on the left in Documents A, B and C and in Document A, it even forms a small loop.

97.  In relation to the character “Teh” （德）, there are these dissimilarities.  (1) In Document A, the vertical stroke in the left radical “彳”extends to a very low position.  (2) In Documents A, B and C, the first horizontal stroke of the right radical is written from left to right with a slight overhand movement, then turning back at the end with another slightly overhand movement, finally coming down and crossing the first horizontal stroke and not as in the sample signatures, with an underhand pen movement; a compressed loop is formed in Document B and C but this does not occur in the sample signatures. (3) In the right radical, the point of intersection between the horizontal stroke and the vertical stroke is towards the left end of the horizontal stroke and not in the middle. (4) In Document B, the last dot on the left in the right radical is in the shape of a slant reaching lower than the lowest part of the curve which is next to it and not as part of the pen movement. (5) In Documents B, C and D, in the lower part of the right radical, the curve in the radical “心” is in an oval shape which is almost closed and not open and rounded as in the sample signatures. 

98.  In relation to the character “Huei” （輝）, there are these dissimilarities.  (1) In Document A, the slant on the left radical “光”is very low reaching when compared with the samples.  (2) The hook which is the last stroke of the left radical “光”forms a very narrow and sharp V-shape and not a U-shape.  (3) In Document B, the first vertical stroke on the left radical “光”comes down slightly from the right to the left and not slightly from left or vertically downwards.  (4) In Document D, there is a connection between the left and right radical which does not appear in all the samples.  (5) The last long vertical stroke is either straight down or with its end going towards the lower right and not either straight down or slightly towards the lower left.

99.  In respect of these dissimilarities, I would make the following observations.  First, in the examples given above, while they show a difference with the more recent sample signatures, most of them bear a similarity with either one or both of the 1958 signatures with slight variations: examples (1), (2), (3), (4), (5) and (8) in relation to the character “Wang” （王）; examples (3) and (5) in relation to the character “Teh” （德）; and examples (1) and (2) in relation to the character “Huei” （輝）. 

100.  Secondly, the following points can be noted in relation to the other examples.  In example (6) in relation to the character “Wang” （王）, the “closed loop” in the middle may be due to the use of the broad nib calligraphic pen and the nature of the paper, for example, the loop in Document C is also open but smaller.  Example (1) in relation to in the left radical “彳”in the character “Teh” （德）is a trivial difference: the same vertical stroke in some of the samples also display a long stroke although slightly shorter.  In example (4) in the same character, the dot in the form of a slant also appears in the other three Wang signatures and some of the sample signatures except that the one in Document B is longer.  Example (3) in relation to the first vertical stroke on the left radical “光”in the character “Huei” （輝）can be regarded as a trivial difference since the same pen movement also appears in the other three Wang signatures as well as the 1958 signatures and the difference is only a matter of extent.  Example (4) in relation to the same character is a minor difference since the pen movement in both the Wang signatures as well as the sample signatures is similar; it is just that the pen does not leave the paper and continues to start writing the first stroke of the right radical.  Example (5) is a common form when the writing is natural and free flowing. 

101.  Lastly, I would say that these differences are easily detectable and they are not of great significance.

102.  On the other hand, example (7) in relation to the character “Wang” （王）as described in paragraph 96 above can be regarded as a significant dissimilarity.  The timing and the position of the turning of the pen in writing the middle part of this character are personal characteristics of the writer.  In the Wang signatures, both the timing and the position of the turning of the pen are different from those in the sample signatures, except possibly in sample 668 (one of the 1958 sample signatures) where the difference is not as great as in the others.  This, in my view, is a subtle, inconspicuous and important difference. 

Dissimilarities - Example (7) in relation to the character “Wang” （王）

[Taken from the Appellant’s Case]

(a) signatures in Documents A, B, C and D

(b)signatures in the samples

The way the central part of the character is written.

103.  Similarly, example (2) in relation to the character “Teh” （德）as described in paragraph 97 above can also be regarded as a significant dissimilarity.  There is a difference in the pen movements: in the Wang signatures, both the first horizontal stroke from left to right and the immediate return stroke from right to left are slightly overhand movements but in all the sample signatures (including the 1958 signatures) these two strokes are done with an underhand or horizontal movement.  As I mentioned earlier, a difference in pen movement of the same stroke which is a personal characteristic of the writer’s writing habit may be an important distinguishing feature.  In my view, this is a subtle, inconspicuous and significant difference.  I do not think there is any good explanation for the difference.

Dissimilarities - Example (2) in relation to the character “Teh” （德）

[Taken from the Appellant’s Case]

(a) signatures in Documents A, B, C and D

          Both (1) and (2) are overhand pen movements.

          (b) signatures in the samples

          (1) is underhand or horizontal pen movement; (2) is overhand pen movement;

104.  In my view, all except two of the dissimilarities raised by counsel are not significant; however, the two dissimilarities mentioned in the preceding two paragraphs are indeed important and weighty differences and can be validly taken into account in the equation as indications of forgery.

Matching natural variations

105.  It is accepted that there are bound to be differences displayed among the handwritings by the same person.  These differences are described as natural variations and they are useful in identifying whether a questioned signature is genuine or not.  As Osborn said at pages 368 and 369:

“Another indication of genuineness in a holographic document or a considerable amount of writing, or in two or more disputed signatures, are certain natural variations in the details of the writing.  It is difficult for the inexperienced or unthinking examiner to understand that a certain extent of variation in a group of several signatures and variation in repeated words and letters in a continuous holographic document can be evidence of genuineness.  The forger does not understand this necessity for natural variation and, as nearly as he can, makes words and letters just alike.

It necessarily follows, therefore, that if the several lines of a disputed document, or several signatures under investigation, show these natural variations of writing of the same word or letter, all of course within the scope of variation of the genuine writing, this variation itself, surprising and paradoxical as it may appear, is as strong evidence of genuineness as the opposite condition is evidence of forgery.”

106.  It is important first to identify the natural variations among the sample signatures in order to determine the range of the variation.  If these variations can also be found in the questioned signature, this would be strong evidence of genuineness.

107.  Where there are more questioned signatures than one which are all written by the same person (as in the present case), there would also be natural variations among these questioned signatures.  If the group of natural variations found among several questioned signatures can be compared with the group of natural variations found among the known and genuine signatures, any matching between these two groups of natural variations would be even stronger evidence of genuineness.  A compelling inference can be drawn that the author of the group of questioned signatures is the same person signing the group of genuine signatures as they display similar natural variations. 

108.  There is, however, one important qualification.  Natural variations, whether for comparing with a questioned signature or for matching with the natural variations found in a group of questioned signatures, must be subtle and inconspicuous.  See Osborn, p. 250.  If these natural variations are common or obvious, they would easily be imitated or taken into account by a forger in perpetrating the forgery. 

109.  Counsel for the Wife relies on six examples of matching natural variations to support his contention that the Wang signatures are genuine.  (1) The direction of the first slant or dot in the left radical “彳”in the character “Teh” （德）.  In the Wang signatures, there are two ways of executing this stroke: one from the right to the left and the other from the left to the right.  This variation is present in some of the sample signatures. 

110.  (2) The pen wielding movement at the end of the vertical stroke in the left radical “彳”in the character “Teh” （德）.  In the Wang signatures, there are two ways of ending this stroke: one ending with a heavy pressure and the other with a hook moving upwards to the right.  The same variation can be seen in some of the sample signatures. 

111.  (3) The first two strokes of the right radical in the character “Teh” （德）.  In the Wang signatures, sometimes the two strokes are connected to form a loop but sometimes they are not connected.  This also appears in some of the sample signatures. 

112.  (4) The two dots inside the middle part of the right radical in the character “Teh” （德）.  In two of the Wang signatures, there is only one dot and in the other two signatures, both dots are present.  This is a variation which also occurs in some of the sample signatures. 

113.  (5) The pen wielding movement in writing the “shelter” on top of the right radical in the character “Huei” （輝）.  The questioned signatures show two ways of doing this: one is a smooth curve and the other is a wavy curve.  This variation is present in some of the sample signatures. 

114.  (6) The connecting movement in the middle of the right radical “軍”in the character “Huei” （輝）.  There are three ways of executing this movement: one in the shape of the numeral “2”, another one in the shape of a heart pointing to the right, and the third one also in the shape of a heart but pointing slightly upwards to the right.  All three ways are also present in some of the sample signatures. 

115.  In dealing with these matching natural variations, Yam J adopted the Father’s written submissions and considered that the matching natural variations could be explained by the theory that the questioned signatures were simulated from more than one model.  This cannot be right.  There is absolutely no evidence in support of this theory.  This line of reasoning also shows that the trial judge had misunderstood the significance of matching natural variations as a very powerful means of identifying the genuineness of a group of questioned signatures.  Yuen JA seemed to accept the multi model theory since she said she saw no reason to assume that the person who forged these signatures had access to only some of the known signatures.  Like Yam J, she also misunderstood the purpose of the exercise of matching natural variations.  In any event, both Yeung and Yuen JJA rejected these variations as being common and not unique to the writer and also easily discoverable.

116.  In my view, these examples are telling.  While examples (1), (2), (3) and (5) are indeed either too common or obvious and so, do not carry very strong weight in the balancing exercise, examples (4) and (6) are different.  The fact that the same natural variations appear in both the Wang signatures and the sample signatures cannot be a mere co-incidence.  These two matching natural variations are quite subtle, inconspicuous and not easily noticeable.  They are personal characteristics in Mr Wang’s writing.  These matching variations are very convincing features pointing towards genuineness.

Matching natural variations – Example (4)

[Taken from the Appellant’s Case]

Matching natural variations – Example (6)

[Taken from the Appellant’s Case]

          The 3 different pen movements in the middle.

117.  It would seem that the trial judge and the majority of the Court of Appeal had overlooked the usefulness of the matching exercise and had failed to take into account these two important examples of matching natural variations which are very strong indications of genuineness. 

Inherent probabilities/improbabilities of forgery

118.  The authenticity of all the 1990 documents is challenged and they stand or fall together.  If they are all forgeries, that would mean that whoever was behind this must have simulated by one way or another at least eight signatures (four Wang signatures and four Tse signatures), not counting the one by a Mr Lee in Document A.  This leads to the argument as to why a forger would have done this.  Why would he or she have taken such an additional risk?  This would only expose more documents to scrutiny.

119.  Furthermore, three of the documents including the most important document, Document A, were written on flimsy paper.  While the Tse’s signatures were written in ball pen ink, the Wang signatures were written with a calligraphic pen with a broad nib.  As I have discussed earlier, such instrument would reduce the writing speed and increase the difficulty in writing, not to mention simulation.  As a result of the use of such pen and paper, there were scratches and cuts in the paper.  It can be strongly argued that a forger would have exercised greater care.  Why would he or she risk damaging the paper as it did by using such writing instrument and paper?  The least he or she would do was to do it in such a way so as to minimize the risk of being exposed and not to draw attention to any inadequacies in the forgery. 

120.  In my view, these matters tend to show the inherent improbabilities of a forgery.  The judge had not fully appreciated their relevance and importance.

Consolidated evaluation

121.  It can be seen from the above discussions that in relation to the handwriting evidence, the judge had fallen into several errors.  First, in the absence of clear expert evidence, he was not entitled to hold that there was a difference in orientation and alignment in the Wang signatures.  Secondly, in the presence of expert evidence to the contrary, he should not have held that there was retouching in the Wang signatures.  He had also wrongly regarded the presence of tremors as a sign of unnaturalness when the experts did not consider them to be of great significance.  As a result, he was wrong to hold that the Wang signatures were not written naturally.  Thirdly, in considering whether there were any good reasons for the close resemblance between the Wang signatures and the 1958 signatures, the judge had failed to appreciate the relevance and importance of the nature of the paper on which the Wang signatures were written and the use of the calligraphic pen.  Fourthly, the judge had attached too much weight to the differences but failed to attach any or sufficient weight to the naturalness of the writing and the matching natural variations.  Fifthly, he had failed to make a consolidated evaluation of all the handwriting evidence.  The majority of the Court of Appeal had not corrected these errors.  As a result of these errors, the conclusion of the judge and the majority of the Court of Appeal that the Wang signatures were forgeries was fundamentally flawed and cannot stand. 

122.  It is necessary to consider all the handwriting evidence in the round in a consolidated evaluation.  No doubt all the circumstances and all the similarities and dissimilarities between the Wang signatures and the sample signatures must be taken into consideration.  Some of the similarities and dissimilarities are superficial and insignificant.  However, there is clear evidence that the Wang signatures were written entirely naturally and in a free flowing style.  This is a strong indication against forgery.  Two important matching natural variations also strongly suggest that these signatures are genuine.  On the other hand, the Wang signatures resemble the 1958 signatures but look different from the other sample signatures.  The explanations given for such resemblance are not entirely satisfactory.  This coupled with the two significant dissimilarities are strong indicia against a finding of genuineness.  Bearing also in mind the inherent probabilities pointing away from a conclusion of forgery, I must say that all the handwriting evidence when considered together is quite inconclusive.  It cannot be said with any degree of confidence that a finding of genuineness is any more likely than a finding of forgery or vice versa. 

Unfair criticism of Professor Jia 

123.  I would not like to end my discussion of this part of the evidence without commenting on the criticisms made by the trial judge on Professor Jia. 

124.  The professor is a handwriting expert trained in the Mainland.  I need not repeat his qualifications and experience.  Suffice it to say, he is a well-known expert in this field.  The judge is of course entitled to accept or reject the professor’s opinion.  This depends on how sound and convincing the reasons given by Professor Jia are in support of his conclusions.  However, the judge described him as a “hired gun”.  Furthermore, the judge had twice in his judgment remarked that the professor was dishonest.  These are very harsh comments and serious accusations to make on and against an expert, unless there are very good grounds to do so.  In my view, there are none.

125.  The judge’s criticisms were mainly based on the premise that Professor Jia said he could not see on the microscope what the judge said he could see.  As it turns out, this could well be a misunderstanding as there was the possibility of optical illusion when things are seen enlarged many times under the microscope.  I should point out that counsel for the Father did not seem to support these remarks.  What is more, the judge did not seem to have put to the professor the judge’s own misgivings on the evidence or the conduct of the professor, so that he could have an opportunity to explain or defend himself.  Counsel for the Father accepted part of the responsibility and offered an apology.

Conclusion

126.  Mr Tse said clearly and unequivocally in his statements that he saw Mr Wang sign on the 1990 documents and then he put his signatures there as a witness.  Although this could not be tested in cross-examination since Mr Tse had passed away, however, having considered all the evidence including the handwriting evidence which is inconclusive and equivocal, there is no good reason not to accept Mr Tse’s evidence as true.  That, in my view, puts an end to the case for the Father.  For the reasons given in this judgment and those given by Mr Justice Ribeiro PJ, I would allow the Wife’s appeal. 

Mr Justice Ribeiro PJ:

A.      Introduction

127.  In this appeal, the contest between the parties concerns the genuineness of a will propounded by the appellant.  The approach adopted by the courts below to the assessment of the evidence calls for examination in the light of the applicable principles.  Also to be addressed is the appellant’s contention that the nature of the judgment at first instance demonstrates that she was deprived of a fair trial.

128.  On 10 April 1990, Teddy Wang Teh Huei (“Mr Wang”), the appellant’s husband, was kidnapped.  He has never since been seen.  On 12 April 1997, the respondent, Mr Wang’s father, applied to the court for leave to swear to the death of Mr Wang.  The appellant resisted the application, contending that Mr Wang was still alive.  After a series of procedural disputes, the substantive application came on for a hearing before Yam J on 8 September 1999.  The appellant gave evidence of certain contacts she had had with the presumed kidnappers, including an incident in 1996, as the basis of her belief that Mr Wang was still alive.  However, her evidence was rejected and, on 22 September 1999, Yam J granted the respondent leave to swear to his belief that Mr Wang’s death had occurred on or since 10 April 1990. 

129.  This enabled the respondent to commence the present probate proceedings, which he did on 24 September 1999.  Naming the appellant as the defendant, he claimed probate in solemn form of a will dated 15 March 1968 (“the 1968 Will”).  Under the 1968 Will, the respondent is named the sole beneficiary of Mr Wang’s estate.  As Mr Wang had enjoyed great wealth, the estate is very substantial.  While in the pleadings, the authenticity of the 1968 Will was not admitted and while one of the attesting witnesses was called at the trial to prove that will, it could not seriously be disputed that it was indeed the will of Mr Wang. 

130.  The proceedings leading to the present appeal were therefore effectively concerned only with the appellant’s counterclaim for probate in solemn form of a will purporting to be the will of Mr Wang, written in Chinese and dated 12 March 1990 (“the 1990 Will”).  If valid, it revokes the 1968 Will.  Under the 1990 Will, the appellant takes as sole beneficiary of Mr Wang’s estate.  The respondent alleges that the 1990 Will is a forgery.  Whether Mr Wang did in fact sign the 1990 Will was the ultimate factual issue at the trial.

B.      The factual background

131.  The respondent was born in 1911.  Mr Wang was his eldest son, born in 1934.  There were two other sons and two daughters.  The appellant was born in 1937.  She and Mr Wang had known each other in childhood.  They got married in Hong Kong in 1955 when the appellant was 18 and Mr Wang was 21.  By that time, the respondent, accompanied by Mr Wang, had moved to Hong Kong from Shanghai.  The respondent established a business of trading in chemicals and, in 1958, went into the business of importing plastics with the participation of his nephew Yih Lee Kong (“Mr Yih”) as well as Mr Wang. 

132.  In 1960, when aged 26, Mr Wang made a will (“the 1960 Will”) which was later revoked by the 1968 Will.  It was typed in English, dated 23 April 1960 and prepared by Messrs P C Woo & Co, a firm of solicitors which also provided the two attesting witnesses.  By the 1960 Will (which was signed using his English signature) Mr Wang left his entire estate to the appellant and the respondent as tenants in common in equal shares, naming them as joint executors.  It also provided that if Mr Wang and the appellant were to have any children, she was to hold her share on trust for such children.  As it turns out, there were no children of their marriage.

133.  It was also in 1960 that Mr Wang began his involvement in the property development business which was to prove extremely successful.  He established Chinachem Investment Company Limited (“Chinachem”) for this purpose.  Shares in that company were allotted, among others, to the appellant, the respondent and Mr Yih.  Mr Wang was the largest and controlling shareholder. 

134.  In 1968, serious matrimonial problems developed between Mr Wang and the appellant.  It appears that he believed that she had had an adulterous affair.  This evidently led to his making the 1968 Will whereby the appellant was excluded and his entire estate was left to the respondent.  He was then aged 33 and the respondent aged 56.

135.  Like the 1960 Will, the 1968 Will was typed in English and prepared by solicitors, this time by Messrs F Zimmern & Co.  It was signed by Mr Wang using his English signature and witnessed by two partners in that firm, namely Mr Donald Quintin Cheung (“Donald Cheung”) and Mr Frederick Zimmern.  The former gave evidence at the trial as an attesting witness. 

136.  Mr Wang also took steps to have certain assets held by the appellant transferred to himself and, as evidenced by his letter dated 7 July 1969 to the Hong Kong and Shanghai Bank, he withdrew “for the time being” a power of attorney he had given to her.  Photographs later found indicate that Mr Wang himself may have had liaisons with other women during this period.

137.  Earlier, on 12 February 1968, Mr Wang had rented a safe deposit box from the Bank of Canton.  He named Mr Yih as his agent, authorising him to open it and, according to the respondent, he told Mr Yih to open the box if requested to do so by the respondent.  Mr Wang had one of the keys and the other was held either by Mr Yih or by the respondent, but was said at the trial to have been lost.  It was in this safe deposit box that the original of the 1968 Will was later found. 

138.  It is plain, however, that Mr Wang and the appellant later became fully reconciled.  There had apparently been some friction between Mr Yih and the appellant but at the end of January 1970, Mr Yih left the business, selling his shares in Chinachem to Mr Wang.  From then onwards, the appellant assumed an increasingly important role alongside Mr Wang in the operation of their expanding and highly successful property development business.  The respondent, who had gone into semi-retirement in 1965, became fully retired in 1977.

139.  In 1983, the appellant and Mr Wang were both kidnapped.  The kidnappers released the appellant so that she could make ransom arrangements.  She did so and secured the release of Mr Wang on payment of a ransom of $75 million.

140.  After the kidnapping, Mr Tse Ping Yim (“Mr Tse”) came to work as a manservant for the Wang couple.  From 1983 onwards, he resided at their home, accompanying them to and from work and rendering various personal services for them.  As indicated below, his purported signature appears on the 1990 Will. 

141.  Beginning in about 1987, professional advisers were retained with a view to drawing up a tax shelter plan for the assets of Mr Wang and the appellant.  It was apparently envisaged that their assets would be placed in an offshore trust to be set up with the couple as the only intended beneficiaries.  However, having received advice that further consideration had to be given to the offshore aspects of the plan, the appellant instructed their advisers on 15 February 1990 to suspend their efforts for the time being.  In the events which occurred, the plan was never brought to fruition.

142.  On 10 March 1990, Mr Wang, who was a keen rider, injured himself when he fell off a horse at the Jockey Club at Beas River.  He suffered a concussion and was hospitalized for two days.  He discharged himself against medical advice on 12 March 1990, which is the date of the 1990 Will.  According to statements made by Mr Tse (discussed further below), Mr Wang had gone back to the office upon leaving the hospital and had signed the 1990 Will on that very day, with Mr Tse signing as a witness.  The respondent alleges that Mr Tse’s statements are false.

143.  On 10 April 1990, Mr Wang was kidnapped for a second time.  The appellant was contacted by the kidnappers who demanded a ransom of US$60 million.  She made payment of about half that amount but before further payments could be made, the kidnappers broke off contact and Mr Wang’s release was not secured.  As noted above, he was never seen again.  Some time after the kidnapping, the appellant resumed active control and management of the Chinachem group of companies which has since continued to enjoy conspicuous financial success. 

C.      The proceedings prior to trial 

C.1    Application to swear death and other preliminary proceedings 

144.  The probate proceedings instituted by the respondent had a false start.  Having issued an application for leave to swear to Mr Wang’s death on 12 April 1997, those proceedings were discontinued on 2 September 1997 after the respondent commenced a probate action (HCAP 6/1997) on 28 May 1997 seeking a declaration that Mr Wang was dead and claiming probate in solemn form of the 1968 Will. 

145.  However, the Court of Appeal ruled that the probate action could not proceed without first obtaining the court’s leave to swear to Mr Wang’s death.  HCAP 6/1997 was therefore struck out and the respondent issued a fresh application for leave to swear death (in HCMP 79/1999), obtaining such leave on 22 September 1999.  The present proceedings (HCAP  8/1999) were started on 24 September 1999.  Additionally, he commenced proceedings on 12 June 1999 for the appointment of interim receivers over Mr Wang’s assets (HCMP 3454/1999).

C.2    The appellant’s affirmations of testamentary scripts 

146.  On 18 November 1999, the appellant filed her affirmation of testamentary scripts in the present probate proceedings.  It confirmed an earlier version filed in the abortive probate proceedings and added a reference to certain informal wills which became known in the litigation as the “holiday wills”.  Such an affirmation is required by O 76 r 5 to be filed by all parties to a probate action.  It must describe any testamentary script of the deceased person, whose estate is the subject of the action, of which the deponent has any knowledge.  The term “testamentary script” includes a will or draft will, written instructions for a will and any document purporting to be evidence of the contents of a will said to have been lost or destroyed.  All testamentary scripts in the possession or under the control of the deponent are required to be lodged in the High Court Registry.  With her affirmation, the appellant lodged two sealed envelopes.  One was later found to contain the 1990 Will (as well as other documents referred to below as “the 1990 documents”) and the other, a copy of the 1968 Will.

147.  In her affirmations of testamentary scripts the appellant declared:

“In about February or March 1990 my husband made a new will and a few days later he gave me a sealed envelope which he told me contained the will and various confidential instructions relating to his estate.  He gave me this envelope for safe-keeping and I have kept it ever since in my safe.  When he gave me the envelope he told me that it should not be opened until after his death.  Out of respect for his wishes I have never opened the envelope.”

She added:

“On the basis of what Mr Wang told me when he gave me the envelope in 1990 I believe that the documents in the envelope refer to highly private and confidential matters relating to the disposition of my husband’s estate after his death.  I strongly believe that that information should not be disclosed until (he) is dead or declared dead by this court.” 

The evidential status and effect of her affirmations will require further consideration below.

148.  As previously indicated, the appellant’s stance had consistently been that, so far as she was concerned, Mr Wang was still alive and that nothing should be done in the course of the litigation which might jeopardize his safety.  On 29 April 1998, Yam J acceded to her application for a direction that the envelopes she had lodged in court should remain unopened unless or until a declaration of Mr Wang’s death was made.  The envelope accordingly remained in court unopened until the appellant changed her mind and agreed to it being opened by Yam J on 6 September 1999.

C.3    Opening the sealed envelope 

149.  It was in the afternoon of 6 September 1999 that the contents of the two sealed envelopes lodged in court by the appellant were revealed.  Nothing further needs to be said about the copy of the 1968 Will which was in one of the envelopes.  In the other envelope were found four smaller envelopes, each containing a single document.  These documents were marked by Yam J in turn as Documents A, B, C and D respectively (“the 1990 documents”) .  The 1990 Will was marked as Document A.  The order in which each document was marked was fortuitous.

150.  The solicitors representing the parties at that hearing were allowed to take copies.  Present in court on that day were Mr Allan Leung Chun Yue (“Allan Leung”) of Messrs Lovell White Durrant (as the firm was then known), solicitors then acting for the appellant in the proceedings, as well as Mr Joseph W K Poon (“Joseph Poon”) of Messrs Poon, Woo & Johnny T K Cheng, described as a consultant to the appellant.

151.  Seeing that the 1990 documents had apparently been legibly signed by Mr Tse, Joseph Poon arranged for a statement to be taken from Mr Tse that evening.  This was done by Mr Alex Lam Fong Vai (“Alex Lam”), a solicitor in his firm.  It took the form of a statutory declaration.  The circumstances in which that statement was taken, as well as the circumstances in which a second statement (in the form of an affirmation) was taken by Allan Leung three days later, on 9 September 1999, will require further consideration below. 

D.      The 1990 documents 

D.1    Document A – the 1990 Will 

152.  In her affirmations of testamentary scripts the appellant did not specify the exact date when the sealed envelope was handed to her by Mr Wang.  However, if the will is genuine, it would follow that she must have received it a few days after 12 March 1990. 

153.  The 1990 Will was homemade (in the sense that it was not professionally prepared) and handwritten.  It bears three signatures.  First, it purports to be signed at the bottom left margin of the document by Mr Wang using his Chinese signature (the Chinese characters being 王德輝).  Secondly, it appears to be signed in Chinese halfway down the right margin of the document by Mr Tse as an intended attesting witness (the characters being 謝炳炎).  Thirdly, it bears a signature in English which is not fully legible, positioned beneath the purported signature of Mr Tse.  It was also apparently intended as the signature of an attesting witness and looks as if it is the signature of someone surnamed Lee (“the Lee signature”). 

154.  It is the appellant’s case that while she cannot assert as a fact the identity of the author of the Lee signature, she considers it to resemble signatures in her possession signed by a Mr Lee Kin Sang (“Lee Kin Sang”) who had died in 1996.  Lee Kin Sang was a construction contractor well known to Mr Wang, having had dealings with Chinachem and having occupied an office downstairs in the same building.  Discovery was given by the appellant of documents bearing known signatures of Lee Kin Sang but no positive case was advanced that he had signed the 1990 Will.

155.  The text of Document A which constitutes the 1990 Will, in its certified translation, reads as follows:

“Will.  12th day of March 1990, I, Wang Teh Huei, of 15 Bluff Path, The Peak, Hong Kong, solemnly make this will that after my death, all my property shall be bequeathed to my wife Nina Kung”.

156.  On the appellant’s case, only Document A is testamentary in character, the other documents being merely precatory or indeed, of no operative relevance in the testamentary context.  It may be noted that only Document A bears two signatures additional to Mr Wang’s.  The other three were signed only by Mr Wang and Mr Tse.

D.2    Document B 

157.  Document B is also a document handwritten in Chinese.  Again, it purports to be signed by Mr Wang at the lower left margin of the document and by Mr Tse half-way down the right margin.  In its certified translation it reads:

“After the death of me Wang Teh Huei, all my property shall be left to my wife Nina Kung for her management and no one shall disagree with that.  I love my wife.  She is my dearest in the world.  After my death, all of my property, real estate and my body shall all belong to my beloved wife.  Although my father and mother disappoint me, I still insist that my wife must take care of them and also that ailing younger sister who can never possibly support herself.” 

D.3    Document C 

158.  Document C is another document handwritten in Chinese.  It also bears purported signatures of Mr Wang and Mr Tse in similar positions to those described above.  In its certified translation it reads:

“After my death, all my property shall be left to my wife and no one shall object to that. My wife shall manage all the property, but shall never deliver or give any monetary benefit or property to any others of my Wang’s family. They are all disappointing. But, neither shall (my) wife share out the property to anyone of your Kung’s family, because everyone of your Kung’s family is also disgusting.”

D.4    Document D 

159.  Unlike the other documents, the short text of this document is handwritten in English, although it bears Chinese signatures purporting to be those of Mr Wang and Mr Tse, similarly positioned as in the other documents.  The text consists of the four words “One life one love” arranged on the page in four downward diagonal steps from left to right.

D.5    Features of the 1990 documents 

160.  It is common ground that the 1990 documents have the following features (among others that are no longer material):

(a) All the purported signatures of Mr Wang were signed using a broad-nibbed calligraphic pen.

(b) Those of Mr Tse were signed using an ordinary ball-point pen.

(c) An image or impression of the words “one life one love” written as on Document D appears faintly on Document C.

(d) An image or impression of the words “one life one love” in a horizontal line, with the word “Nina” just beneath, appears on Document B and more faintly on Document A.  The inference is that such image or impression resulted from someone writing those words on another document (not included among the 1990 documents) superimposed on Documents A and B.  That presumed missing document was referred to as Document E at the trial.

(e) Documents A, B and C are written on thin paper and Document D is written on thicker paper.

(f) The writer of the texts of Documents A, B and C was not Mr Wang.

(g) The writer of the text of Document A was not the same person as the writer of Documents B and C.

(h) Documents B and C were probably written by the same person.

E.      The trial and the decision of the Court of Appeal 

161.  The trial commenced before Yam J on 6 August 2001.  It was to last for 172 days over a 14-month period, concluding on 15 October 2002.  The respondent (who opened as plaintiff in the proceedings) called Donald Cheung as a witness attesting the 1968 Will.  The respondent himself gave evidence as to the provenance of the 1968 Will as well as evidence of a general nature on the family background and his relationship with Mr Wang and the appellant.  He called certain witnesses, including Mr Yih and Mr Wang’s sister, Ms Teresa Tak Shyan Sun (“Teresa Wang”), to give circumstantial evidence as to Mr Wang’s character and habits intended to cast doubt on the genuineness of the 1990 Will.  He adduced handwriting evidence to the same end, the experts instructed on his behalf being Mr Gus Lesnevich (“Mr Lesnevich”) and Mr David Tsui Chee Keung (“Mr Tsui”).  An independent handwriting expert, Mr Patrick Cheng Yau Sang (“Mr Cheng”), head of the Questioned Document Section in the Government Laboratory, was also called by him.  Furthermore, he called evidence as to the date of the ink used in writing the signatures but Mr Erich J Speckin, the witness claiming to be able to effect such ink-dating, was wholly discredited.  Ink-dating was not thereafter pursued. 

162.  The appellant, for her part, chose not to give evidence.  Unfortunately, by the time the trial started, Mr Tse was no longer alive.  He had died of liver cancer on 6 December 1999.  However, the appellant relies on the statements he had given to solicitors regarding execution of the 1990 Will by Mr Wang (“the Tse statements”).  Alex Lam and Allan Leung were called to testify on the taking of those statements.  The appellant also relies on her affirmations of testamentary scripts as evidence of the provenance of the will.  She called several witnesses to give evidence as to the nature of the relationship, both business and personal, between herself and Mr Wang and as to the character and habits of Mr Wang.  These included Mr Anthony Quintin Cheung Ngan Kwan (“Anthony Cheung”), Donald Cheung’s son and godson to the Wang couple;  Mr Cecil Chao Sze Tsung (“Cecil Chao”), who had entered into numerous property development joint ventures with the Wangs; and Mr Edwin Cheung Hon Kit (“Edwin Cheung”), a former employee of Chinachem.  Testifying as the appellant’s expert on handwriting was Professor Jia Yu-Wen (“Professor Jia”) who works for the Criminal Police College of China in Shenyang.  She also called her brother Dr Kung Yan Sum (“Dr Kung”) in connection with what were referred to at the trial as “holiday wills”.

163.  Judgment was delivered on 21 November 2002 in which Yam J dismissed the appellant’s counterclaim and granted a decree of probate in solemn form in respect of the 1968 Will.  He also ordered that 85% of the costs should be paid by the appellant on an indemnity basis.

164.  On 28 June 2004, after a hearing which lasted 28 days commencing on 29 September 2003, the Court of Appeal (CACV 460/2002, Yeung and Yuen JJA and Waung J), dismissed the appellant’s appeal by a majority, with Waung J dissenting.

F.      Concurrent findings and the present appeal 

165.  Where, as in the present case, the trial judge and the Court of Appeal (whether unanimously or by a majority) have made concurrent findings adverse to the appellant, the principles adopted by this Court in Sky Heart Ltd v Lee Hysan Co Ltd (1997-98) 1 HKCFAR 318 apply.  The Court generally refuses to entertain an appeal against such findings unless there has been some miscarriage of justice or violation of some principle of law or procedure in the sense described in that judgment (at pp 333-338).

166.  In the present case, the grounds of appeal involve a challenge to the basis of the findings themselves.  The appellant argues that those findings were reached by applying erroneous principles of law or procedure such that those findings cannot stand, satisfying the Sky Heart requirements.  As those contentions were plainly seriously arguable, the Court considered it appropriate to hear the substantive appeal rather than to exclude the challenge on the basis of concurrent findings.

G.      The legal principles

G.1    The Wills Ordinance, Cap 30

167.  At the date of the 1990 Will, section 5 of the Wills Ordinance materially provided as follows:

Signing and witnessing will

(1) Subject to subsection (2), no will shall be valid unless it is in writing and executed in accordance with the following rules:

Rule 1.    It shall be signed at the foot or end thereof by the testator, or by some other person in his presence and by his direction.

Rule 2.    Such signature shall be made or acknowledged by the testator in the presence of two or more witnesses present at the same time, and each witness shall attest by his signature the signature of the testator, or of the person signing for him, in the presence of the testator, but no form of attestation shall be necessary.

......

(2) Any will of a Chinese testator written wholly or substantially in Chinese and signed by the testator shall be valid and duly executed although not executed in accordance with the rules set out in subsection (1).

168.  As Mr Wang qualified as a Chinese testator and as the 1990 Will is written wholly or substantially in Chinese, s 5(2) applies.  The will would therefore be valid and duly executed if it was in fact signed by Mr Wang. 

169.  The validity of such a will, unlike non-Chinese wills, was not conditional upon attestation by two or more witnesses.  Nevertheless, as has been noted, the 1990 Will purports to have been witnessed by Mr Tse and the author of the Lee signature.  Where, as in the present case, the central issue is as to whether Mr Wang did in fact sign the will, the evidence of any such attesting witness would obviously be of the first importance, even though attestation was not necessary as a matter of law.

170.  The requirements laid down by s 5(1), and in particular those concerning formal validity, although inapplicable in the present case, are nonetheless relevant to an understanding of the authorities which influenced the approach adopted by the courts below. 

G.2    The burden of proof when a will is propounded

171.  A person who propounds a will has the legal or persuasive burden of satisfying the court that it is the will of the deceased: Barry v Butlin (1838) 2 Moo PC 480; Robins v National Trust Company Limited [1927] AC 515; Wintle v Nye [1959] 1 WLR 284 (HL).

172.  The standard of proof is the balance of probabilities.  The proponent of the will must show on a preponderance of the evidence that it is the will of the deceased: Worth v Clasohm (1952) 86 CLR 439, 453; Fuller v Strum [2002] 1 WLR 1097, 1120.

173.  This means that the proponent of the will has the persuasive burden of proving on a balance of probabilities that:

(a) there was due execution of the will;

(b) the testator was of testamentary capacity; and,

(c) the testator knew and approved of the contents of the will.

(Barry v Butlin (above); Harmes v Hinkson [1946] 3 DLR 497 (PC))

174.  In the present case, the dispute is as to whether Mr Wang executed the 1990 Will.  There is no dispute as to Mr Wang’s testamentary capacity nor as to whether he knew and approved of the contents of that will (except in the sense that, if a forgery, it obviously had nothing to do with his testamentary dispositions at all).  However, as appears in what follows, certain authorities on the requirement of knowledge and approval were misapplied in the courts below. 

175.  Where one is concerned with non-Chinese wills, proof of “due execution” requires proof that the will was signed by the testator (or by some other person in his presence and by his direction) and that the statutory requirements for the formal validity of wills (set out in s 5(1) above) were complied with.  Thus, a challenge to due execution might involve not only the question whether the testator had in fact signed the will, but also whether he did so in the joint presence of the attesting witnesses or whether those witnesses had each signed in his presence.  But as we have seen, by virtue of s 5(2), “due execution” in respect of eligible Chinese wills (and in the present case) is established simply by showing “execution” or the fact of signature of the testator.  There is accordingly in our case effectively no difference between “due execution” and “execution”.

176.  If someone wishes to dispute the validity of a will on the grounds that there is want of due execution, or of testamentary capacity, or of the requisite knowledge and approval, that person bears an evidential burden of putting the relevant ground of challenge in issue.  If the evidence adduced by him or otherwise arising in the case is of a sufficient cogency to raise such an issue, the court, when assessing the evidence as a whole at the end of the case, decides whether the proponent of the will has discharged the persuasive burden in relation to the relevant fact in issue on the balance of probabilities. 

177.  On the other hand, where a person seeks to challenge a will on the grounds that the testator was induced to make the will by fraud or by undue influence, such person bears the persuasive burden of establishing the fraud or undue influence: Boyse v Rossborough (1857) 6 HLC 2; Craig v Lamoureux [1920] AC 349 (PC).  These pleas presuppose execution of the will by the testator.  The allegation is that such execution was procured by acts of fraud or undue influence performed by others: Tyrrell v Painton [1894] P 151 at 157; C H Sherrin et al, Williams on Wills (Butterworths, 8^th Ed) §5.9.

178.  The person propounding the will has no burden of disproving fraud or undue influence, although he is obviously likely in practice to adduce whatever evidence may be available to counter those allegations.  Where those issues are raised (assuming that the court is satisfied as to due execution, testamentary capacity and knowledge and approval), the court asks itself at the end of the day whether, having regard to all the evidence adduced, the person opposing the grant has satisfied the court that it is more likely than not that the testator was induced to make the will by fraud or undue influence, as the case may be. 

G.3    The respondent’s case on forgery 

179.  The 1990 Will is propounded by the appellant in her counterclaim.  In response, the respondent pleads as follows:

“......the [respondent] specifically:

(a) denies that the alleged Will contained in the document dated 12th March 1990 together with the other alleged undated testamentary documents (‘the 1990 documents’) were made by Wang Teh Huei as alleged or at all;

(b) in particulars [sic] avers that the signatures of Wang Teh Huei and Tse Ping Yim on the 1990 documents were forged.” 

180.  The respondent makes no allegation of fraud or undue influence and therefore bears no persuasive burden to prove either ground of challenge.  He bears the evidential burden of adducing evidence sufficiently cogent and probative to raise the issue of forgery with a view to defeating the appellant’s claim.

G.4    The relevance of Re H to the respondent’s case on forgery

181.  There was much discussion as to the extent to which the principle explained in Re H (Minors) (Sexual Abuse: Standard of Proof) [1996] AC 563, is applicable in the present case in relation to the respondent’s allegation of forgery. 

182.  The majority in the House of Lords in Re H held that the civil standard requiring proof on a balance of probabilities continues to apply where, in civil proceedings, an allegation is made of criminal (or similarly serious) misconduct, but explained that such standard is to be applied flexibly, factoring in the inherently greater improbability of serious misconduct as compared with lesser forms of misconduct, and therefore requiring the person bearing the burden of proving the allegation to prove it with evidence of a commensurate cogency.  The well-known passage in the speech of Lord Nicholls of Birkenhead states as follows:

“The balance of probability standard means that a court is satisfied an event occurred if the court considers that, on the evidence, the occurrence of the event was more likely than not. When assessing the probabilities the court will have in mind as a factor, to whatever extent is appropriate in the particular case, that the more serious the allegation the less likely it is that the event occurred and, hence, the stronger should be the evidence before the court concludes that the allegation is established on the balance of probability. Fraud is usually less likely than negligence. Deliberate physical injury is usually less likely than accidental physical injury. A step-father is usually less likely to have repeatedly raped and had non-consensual oral sex with his under age stepdaughter than on some occasion to have lost his temper and slapped her. Built into the preponderance of probability standard is a generous degree of flexibility in respect of the seriousness of the allegation.” (at 586)

183.  Given that Re H is concerned with defining the standard of proof required to discharge a legal or persuasive burden borne by a party, it may be said that Re H is strictly inapplicable where one is concerned merely with the discharge of an evidential burden (such as the respondent’s evidential burden regarding forgery).  It is in the nature of an evidential burden that the person discharging that burden does not need to prove anything, but merely to adduce sufficient evidence to require the court to decide the issue in question.  As the learned author of Cross and Tapper on Evidence (Lexis Nexis, Butterworths, 10^th Ed) states:  “the discharge of the ...... evidential burden proves nothing” (at p 134). 

184.  Nevertheless, in a case like the present, the Re H principle is applicable by analogy, operating not as defining a standard of proof, but imposing a standard of cogency which must be satisfied before evidence is considered sufficient to raise a case (here of forgery and of an associated conspiracy) for consideration by the court.  When weighing up and assessing the probabilities in relation to the evidence adduced by the respondent as evidence of forgery, the court must bear in mind the seriousness of the misconduct alleged, recognizing that it carries an inherent degree of improbability. 

G.5    Drawing inferences of forgery or fraud

185.  A related principle should be applied in tandem.  Where, as in the present case, the court is invited to reach a conclusion of forgery as an inference to be drawn on the basis of circumstantial evidence, any such inference must be properly grounded in the primary facts found.  The court guards against indulging in conjecture under the guise of drawing an inference where the primary evidence does not logically and reasonably justify the particular inference in question.  As the High Court of Australia stated :

“Of course as far as logical consistency goes many hypotheses may be put which the evidence does not exclude positively. ......The difference between the criminal standard of proof in its application to circumstantial evidence and the civil is that in the former the facts must be such as to exclude reasonable hypotheses consistent with innocence, while in the latter you need only circumstances raising a more probable inference in favour of what is alleged. In questions of this sort, where direct proof is not available, it is enough if the circumstances appearing in evidence give rise to a reasonable and definite inference: they must do more than give rise to conflicting inferences of equal degrees of probability so that the choice between them is mere matter of conjecture: see per Lord Robson, Richard Evans and Co Ltd v Astley (1911) AC 674, at p 687.  But if circumstances are proved in which it is reasonable to find a balance of probabilities in favour of the conclusion sought then, though the conclusion may fall short of certainty, it is not to be regarded as a mere conjecture or surmise ...” (Luxton v Vines (1952) 85 CLR 352 at 358, citing the unreported decision in Bradshaw v McEwans Pty Ltd (1951) HCA)

186.  The High Court returned a few years later in Jones v Dunkel (1958-59) 101 CLR 298 at 305, to stress the need for a proper foundation for the inference.  It is not permissible merely to choose what may be considered to be the more likely of two guesses if neither is properly justified by the primary facts found.  While the court was divided as to its application to the facts of that case, it was agreed as to the nature of the principle.  Dixon CJ (who was in the minority), referring to the abovementioned passage from Bradshaw v McEwans, stated:

“But the law which this passage attempts to explain does not authorise a court to choose between guesses, where the possibilities are not unlimited, on the ground that one guess seems more likely than another or the others.  The facts proved must form a reasonable basis for a definite conclusion affirmatively drawn of the truth of which the tribunal of fact may reasonably be satisfied.”  (at 305)

And Kitto J (for the majority) cautioned:

“One does not pass from the realm of conjecture into the realm of inference until some fact is found which positively suggests, that is to say provides a reason, special to the particular case under consideration, for thinking it likely that in that actual case a specific event happened or a specific state of affairs existed.”  (at 305)

187.  In HKSAR v Lee Ming Tee (2003) 6 HKCFAR 336, Sir Anthony Mason NPJ acknowledged the need for such a disciplined approach to the drawing of inferences and in particular for inferences of fraud or serious misconduct to be drawn only where such inferences are compelling.  Dealing with an allegation that senior SFC officers had deliberately and improperly terminated an investigation in order to avoid compromising the standing of the subject of the investigation who was acting as an expert witness in a criminal trial in which the SFC was interested, his Lordship stated:

“...... that conclusion was not to be reached by conjecture nor, as the respondent submitted, on a mere balance of probabilities. It was to be plainly established as a matter of inference from proved facts”. (at §72)

Reflecting the Re H principle he added:

“In the particular circumstances, it was for the respondent to establish as a compelling inference that very senior officers of the SFC had deliberately and improperly terminated the investigation into Meocre Li’s conduct for the ulterior purpose alleged, sufficient to overcome the inherent improbability that they would have done so.” (Ibid.)  

It will be convenient to refer to the principle adopted in these cases as “the Lee Ming Tee principle”.

G.6    The respondent’s case involving Mr Tse and “the subtle form of forgery”

188.  If the respondent’s pleaded case had been confined to a non-admission of the genuineness of the 1990 Will, then the appellant would simply have been put to proof of the genuineness of the will.  However, the respondent positively alleges forgery and expressly pleads that both the Wang and Tse signatures on the 1990 Will are forged.  This, in my view, takes the respondent’s case beyond merely challenging genuineness.  He is not merely putting the appellant to proof of the will but is seeking to discharge an evidential burden which involves raising a definite case of forgery with a view to defeating the appellant’s claim for probate.  In the present context, such a case necessarily implies a criminal conspiracy involving both Mr Tse and the appellant, among others. 

189.  Thus, if, on the respondent’s case, Mr Tse’s signatures were forged notwithstanding his having confirmed their authenticity in a statutory declaration and in an affirmation, there must necessarily have been a conspiracy involving the forger or forgers, Mr Tse (who must be lying when authenticating the forgeries and lying about Mr Wang’s execution of the will), the appellant (who must be lying about Mr Wang handing her the envelope telling her that it contained a will that he had prepared) and (if they be different persons) the authors of the texts of Documents A, B and C (who must have known that they were writing out testamentary dispositions and instructions for Mr Wang without his involvement). 

190.  Although the complexion of the respondent’s case changed somewhat when it reached this Court, it continues to bear similar ramifications.  Yam J had found that both the Wang and the Tse signatures were forged, Mr Wang’s by simulation (that is, by the forger writing out the forgeries free-hand after having practised writing the signature using available models of the author’s true signatures) and Mr Tse’s by tracing (that is, by tracing the genuine signature onto a superimposed sheet of paper).  However, in the Court of Appeal, only Yeung JA upheld the latter finding.  The finding of forgery of the Tse signatures was rejected by both Yuen JA and Waung J.  However, Yuen JA added:

“I should however clarify that this should not be taken as an acceptance of the submission that Tse’s signatures on the documents being propounded were made in 1990.”

191.  Although the statement does not convey any express finding and her Ladyship does not examine its implications, it clearly suggests that while Mr Tse did indeed sign the 1990 Will, he did not do so on 12 March 1990 as a witness attesting Mr Wang’s execution of the will as portrayed by the document on its face and as deposed to by Mr Tse.  That such an implication was in fact unspokenly accepted by Yuen JA follows from the fact that she upheld the finding that Mr Wang’s signatures are forgeries.  She therefore necessarily rejected Mr Tse’s statements confirming his attestation of the 1990 Will. 

192.  The theory so espoused was referred to at the hearing as “the subtle form of forgery”.  It is a convenient term to adopt for present purposes.  Such a theory also necessarily implies a conspiracy involving (for the same reasons as those given above) the forger (of the Wang signatures), Mr Tse, the appellant and the authors of the texts of Documents A, B and C.  It is plain that the evidence sought to be adduced in support either of the original case of forgery or of the subtle form of forgery, implying a criminal conspiracy as each of these cases does, has to be evaluated applying the Re H standardand the Lee Ming Tee principle.

193.  Mr Neville Thomas QC, leading for the respondent before this Court, adopted wholesale the subtle form of forgery case before this Court.  Mr Geoffrey Vos QC, leading for the appellant, submitted that such an unpleaded case is not open to the respondent.  Nonetheless, he acknowledged that the special duty of the court in a probate case to ensure that the testamentary dispositions do reflect the intentions of the deceased, means that if the Court were to be satisfied that sufficient evidence exists to support the existence of such a conspiracy, it would not grant probate, but would order the unpleaded issue to be properly pleaded and tried.  That proposition is in my view correct. 

G.7    Conclusions on burden and standard of proof summarised

194.  The conclusions reached in relation to the applicable principles so far discussed may be summarised as follows:

(a) The appellant, as proponent of the 1990 Will, bears the legal or persuasive burden to prove its execution by Mr Wang on the balance of probabilities.  She has to satisfy the court at the end of the day that it is more likely than not that the 1990 Will was in fact signed by Mr Wang.

(b) The respondent has positively pleaded forgery in circumstances necessarily implying the existence of a broader conspiracy.  He has the evidential burden of adducing evidence probative of such forgery which is to be assessed applying the Re H standard and the Lee Ming Tee principle when considering the probabilities and deciding the probative value to ascribe to such evidence. 

G.8    The presumption of regularity 

195.  Before proceeding to an analysis of the evidence in the light of the applicable principles, it is necessary to deal with two complicating legal issues introduced in the courts below.  The first concerns the presumption of regularity and the second concerns what is sometimes called “the second rule in Barry v Butlin”.

196.  In the Court of Appeal, there was considerable discussion of whether the presumption of regularity, expressed in the maxim omnia praesumuntur rite esse acta, is applicable to proof of execution by Mr Wang, bearing in mind in particular that the case involves a Chinese will.  The appellant (then represented by Mr Michael Thomas SC) argued that where the will produced by its proponent appears on its face to be signed by the testator and to comply with the other requirements of validity, it should be presumed, until the contrary is shown, to have been duly executed by the testator.

197.  Yeung JA held that the presumption is inapplicable in the present case.  He expressed doubts as to whether it ever applies to Chinese wills where the safeguards provided by the requirements of formal validity are absent and “particularly when there is suspicious circumstance coupled with the allegation of fraud” (at §407).  Yuen JA, on the other hand, considered the presumption applicable generally, including in relation to Chinese wills.  She held, however, that because of the conflicting evidence, it applied with “much less force” in the present case (at §§884 and 888).  In his dissenting judgment, Waung J held (at §§1061-1104) that the presumption applies and should be regarded as “strong and tenacious” requiring “clear, positive and reliable” evidence for its rebuttal.  He held that it had not been rebutted.  Although reliance was placed on the presumption in the appellant’s printed case, it did not feature in Mr Vos’s submissions before the Court.

198.  The presumption of regularity is of great practical importance in cases concerning non-Chinese wills.  If, in such cases, it is established that the testator and the attesting witnesses had signed the will, then unless significant evidence exists to raise an issue to the contrary, the court will presume, without requiring affirmative evidence, that they did so in conformity with the formal statutory requirements applicable to such wills.  Thus, for example, where the statute requires the testator to sign in the joint presence of the attesting witnesses, proof of the signatures would lead the court to presume such joint presence of the witnesses at the time their signatures were appended to the will.  Such a presumption is made where the attesting witnesses are dead, or cannot be found or identified, or even where they testify that they cannot remember whether the testator did in fact sign in their joint presence. (Burgoyne v Showler (1844) 1 Rob Ecc 4; 163 ER 945; Wright v Sanderson (1884) 9 PD 149; Harris v Knight (1890) 15 PD 170; In the Estate of CR Phibbs [1917] P 93)

199.  The presumption is rebuttable and differs in its strength depending on the circumstances.  Thus, where the will is itself regular on its face and the attesting witnesses have signed against an attestation clause reflecting the statutory requirements (typically along the lines of: “Signed by the testator in the presence of us who both being present at the same time and in his presence have subscribed our names as witnesses”: see Woodhouse v Balfour (1887) 13 PD 2 at 3), this gives rise to a strong presumption of regularity.  Inclusion of such a clause in the will indicates an awareness of and an endeavour to meet the formal requirements, enabling a reasonable inference to be drawn that those requirements were duly met: Vinnicombe v Butler (1864) 3 Sw & Tr 580; 164 ER 1400; Wright v Rogers (1869) LR 1 PD 678; Wright v Sanderson (1884) 9 PD 149. 

200.  However, it must be emphasised that the presumption is as to regularity or, as a leading textbook states, a presumption “that everything was properly done” and “that the requirements of the statute have been duly complied with”: Williams, Mortimer and Sunnucks, Executors, Administrators And Probate, 18th Ed (Sweet & Maxwell, 2000), §12-25.  It has no meaningful role to play in resolving a dispute as to whether the testator’s signature is or is not genuine.  As Lindley LJ stated in Harris v Knight (above, at 179):

“The maxim, ‘Omnia praesumuntur rite esse acta,’ is an expression in a short form, of a reasonable probability, and of the propriety in point of law of acting on such probability. The maxim expresses an inference which may reasonably be drawn when an intention to do some formal act is established; when the evidence is consistent with that intention having been carried into effect in a proper way; but when the actual observance of all due formalities can only be inferred as a matter of probability. The  maxim is not wanted where such observance is proved, nor has it any place where such observance is disproved. The maxim only comes into operation where there is no proof one way or the other; but where it is more probable that what was intended to be done was done as it ought to have been done to render it valid; rather than that it was done in some other manner which would defeat the intention proved to exist, and would render what is proved to have been done of no effect.”

201.  As this passage points out, the maxim applies where what has been proved to have been done justifies the presumption that other steps which ought also to have been taken were in fact taken, the proven facts inherently supporting such a probability.  This does not arise where the dispute focuses on the genuineness of the testator’s signature.  There is no preceding proven fact which makes it inherently more probable than not that the signature is genuine.  When authenticity is in issue, merely producing a document on which appears a signature resembling the testator’s does not provide any basis for presuming, one way or the other, anything about its genuineness.  The issue must be resolved as a matter of evidence in accordance with the usual rules relating to burden and standard of proof, and not by application of any presumption.

202.  Moreover, the fact that the law generally permits the person propounding the will to rely on a presumption of regularity cannot be called in aid as a matter helping to incline the balance towards a finding of genuineness, when taken with other evidence.  What Lord Reid stated in S v S [1972] AC 24 at 41, in connection with the statutory presumption of legitimacy applies with equal force to the presumption of regularity: 

“Once evidence has been led it must be weighed without using the presumption as a make-weight in the scale ......” 

203.  It is accordingly my view that the presumption of regularity has no relevance in the present case.  This conclusion does not depend on whether the will is or is not a Chinese will.  The point is that in relation to the present dispute, there exists no basic fact, proof of which supports acceptance of genuineness of the will as a presumed fact. 

G.9    The second rule in Barry v Butlin 

204.  The influence of the second complication was much more profound in the courts below.  In Barry v Butlin (1838) 2 Moo PC 480, Parke B set out two rules which have since repeatedly been referred to:

“These rules are two; the first that the onus probandi lies in every case upon the party propounding a Will; and he must satisfy the conscience of the Court that the instrument so propounded is the last Will of a free and capable Testator.

The second is, that if a party writes or prepares a Will, under which he takes a benefit, that is a circumstance that ought generally to excite the suspicion of the Court, and calls upon it to be vigilant and jealous in examining the evidence in support of the instrument, in favour of which it ought not to pronounce unless the suspicion is removed, and it is judicially satisfied that the paper propounded does express the true Will of the deceased.” (at 482-483)

205.  The first rule, which allocates the persuasive burden to the person propounding the will, is not controversial.  However, the second rule, which refers to a requirement that the proponent of the will remove or dispel “suspicious circumstances” has bedevilled the approach to the evidence in the courts below.  An understanding of the scope of that rule (referred to here simply as “the second rule”) is therefore essential.  It should, however, be borne in mind that the line of authority under discussion is concerned with the evidential aspects of the knowledge and approval requirement in relation to non-Chinese wills in the context of the statutory formal requirements.  This judgment is not concerned with the knowledge and approval requirement in relation to unwitnessed Chinese wills.  The approach in the latter class of cases may or may not be the same, this being a matter which does not arise for decision in the present case.

206.  As noted above, the person propounding the will has the burden of proving due execution, testamentary capacity and the testator’s knowledge and approval of the will’s contents.  It is, and was at the time Barry v Butlin was decided, well-established that proof of the first two conditions generally meant that knowledge and approval would be rebuttably presumed without the need to adduce affirmative evidence thereof.  Thus, in Cleare v Cleare (1869) LR 1 P&D 655, Lord Penzance stated:

“That the testator did know and approve of the contents of the alleged will is therefore part of the burthen of proof assumed by everyone who propounds it as a will.  This burthen is satisfied, primâ facie, in the case of a competent testator by proving that he executed it.  But if those who oppose it succeed by a cross-examination of the witnesses, or otherwise, in meeting this primâ facie case, the party propounding must satisfy the tribunal affirmatively that the testator did really know and approve of the contents of the will in question before it can be admitted to probate.” (at 657)

207.  This was also Parke B’s starting-point in Barry v Butlin:

“In all cases the onus is imposed on the party propounding a Will, it is in general discharged by proof of capacity, and the fact of execution, from which the knowledge of and assent to the contents of the instrument are assumed ......” (at 484)

208.  The particular circumstance introduced in Barry v Butlin involved the fact that “a party writes or prepares a Will, under which he takes a benefit”.  The second rule states that such circumstance ought generally to excite the suspicion of the court, so that knowledge and approval is no longer presumed.  Instead, the court demands affirmative evidence which it subjects to a “vigilant and jealous” examination, refusing to pronounce in favour of the will unless satisfied that it expresses the true will of the deceased.  That this is the effect of the second rule had been made clear by Parke B in Baker v Batt (1838) 2 Moo PC 319, a few months before his Lordship delivered judgment in Barry v Butlin:

“There is also another principle upon which the Court below has acted, and which has long prevailed in the Ecclesiastical Courts, which is this, -- that if the person benefitted by a Will, himself writes or procures it to be written, the Will is not void, as it would have been by the Civil Law; but the circumstance forms a just ground of suspicion, and calls upon the Court to be vigilant and jealous, and requires clear and satisfactory proof that the instrument contains the real intention of the Testator.” (at 321)

209.  It is therefore important to note that Parke B was not seeking to create a general rule imposing a burden on the proponent “to dispel suspicious circumstances surrounding a will”.  His Lordship was careful, in Barry v Butlin itself, to stress that the second rule was concerned with proof of knowledge and approval, that it did not automatically apply and that it did not displace the usual rules concerning the burden of proof.  Continuing from the passage cited above, his Lordship stated:

“...... it cannot be that the simple fact of the party who prepared the Will being himself a Legatee, is in every case, and under all circumstances, to create a contrary presumption, and to call upon the Court to pronounce against the Will, unless additional evidence is produced to prove the knowledge of its contents by the deceased.  A single instance, of not unfrequent occurrence, will test the truth of this proposition.  A man of acknowledged competence and habits of business, worth £100,000, leaves the bulk of his property to his family, and a Legacy of £50 to his confidential attorney, who prepared the Will: would this fact throw the burthen of proof of actual cognizance by the Testator, of the contents of the Will, on the party propounding it, so that if such proof were not supplied, the Will would be pronounced against?  The answer is obvious, it would not.  All that can be truly said is, that if a person, whether attorney or not, prepares a Will with a Legacy to himself, it is, at most, a suspicious circumstance, of more or less weight, according to the facts of each particular case; in some of no weight at all, as in the case suggested, varying according to circumstances; for instance, the quantum of the Legacy, and the proportion it bears to the property disposed of, and numerous other contingencies; but in no case amounting to more than a circumstance of suspicion, demanding the vigilant care and circumspection of the Court in investigating the case, and calling upon it not to grant probate without full and entire satisfaction that the instrument did express the real intentions of the deceased.” (at 484-485)

210.  This approach has been adopted in many cases.  In some of these, the need for the court to be persuaded by the proponent that the will expresses the deceased’s true testamentary wishes became referred to as the need to prove “the righteousness of the transaction”.  For instance, in Fulton v Andrew(1875) LR 7 HL 448, Lord Hatherley stated:

“......there is a farther onus upon those who take for their own benefit, after having been instrumental in preparing or obtaining a will.  They have thrown upon them the onus of shewing the righteousness of the transaction.” (at 472)

211.  One should not be misled by such terminology into attributing a wider ambit to the second rule than was intended.  A modern authority helpfully summarising the position is Fuller v Strum[2002] 1 WLR 1097 (CA). 

(a) Having referred to the doctrine of “the righteousness of the transaction” as a salutary one, Peter Gibson LJ continued in the following terms:

“But ‘the righteousness of the transaction’ is perhaps an unfortunate term, suggestive as it is that some moral judgment by the court is required.  What is involved is simply the satisfaction of the test of knowledge and approval, but the court insists that, given that suspicion, it must be the more clearly shown that the deceased knew and approved the contents of the will so that the suspicion is dispelled.  Suspicion may be aroused in varying degrees, depending on the circumstances, and what is needed to dispel the suspicion will vary accordingly.  In the ordinary probate case knowledge and approval are established by the propounder of the will proving the testamentary capacity of the deceased and the due execution of the will, from which the court will infer that knowledge and approval.  But in a case where the circumstances are such as to arouse the suspicion of the court the propounder must prove affirmatively that knowledge and approval so as to satisfy the court that the will represents the wishes of the deceased. All the relevant circumstances will be scrutinised by the court which will be ‘vigilant and jealous’ in examining the evidence in support of the will: Barry v Butlin (1838) 2 Moo PC 480 at 483 per Parke B.”  (at 1107)

(b) Chadwick LJ stressed the continued applicability of the usual rules on the burden and civil standard of proof and the need to understand references to removing suspicions in that light:

“...... there is no basis for an approach that requires, in all cases, that a person propounding a will which he has prepared, and under which he takes a benefit, must satisfy the court by evidence which excludes all doubt -- or by evidence which excludes all reasonable doubt (the standard of proof required in criminal proceedings) -- that the testator knew and approved the contents of the will. The standard of proof required in probate proceedings (as in other non-criminal proceedings) is satisfaction on the preponderance (or balance) of probability. But the circumstances of the particular case may raise in the mind of the court a suspicion that the testator did not know and approve the contents of the document which he has executed which is so grave that, as Viscount Simonds observed in Wintle v Nye [1959] 1 WLR 284, 291, it can hardly be removed.” (at 1121)

(c) Longmore LJ put the point simply as follows:

“The vigilance and jealousy of the court is directed to being satisfied that the testator did know and approve the contents of his will; no less but also no more.”  (at 1123)

212.  As previously noted, the testator’s knowledge and approval of the contents of the will were not in issue in the present case.  An attempt by the respondent to raise it as a distinct issue in the pleadings as required by O 76 r 9(3) was rejected by the trial Judge.  However, as discussed below, Yam J (and in a modified way, the majority in the Court of Appeal) evidently did not regard the second rule as a principle confined to laying down more stringent evidential requirements in respect of the testator’s knowledge and approval in the circumstances mentioned. 

213.  The erroneous notion that the second rule is authority for a general requirement that the proponent of a will “dispel all suspicious circumstances surrounding the will” is traceable to a misunderstanding of a passage in the judgment of Lindley LJ in Tyrrell v Painton [1894] P 151 at 157, which states as follows:

“The rule in Barry v Butlin (1838) 2 Moo PC 480, Fulton v Andrew (1875) LR 7 HL 448, and Brown v Fisher 63 LT 465, is not, in my opinion, confined to the single case in which a will is prepared by or on the instructions of the person taking large benefits under it, but extends to all cases in which circumstances exist which excite the suspicion of the Court; and wherever such circumstances exist, and whatever their nature may be, it is for those who propound the will to remove such suspicion, and to prove affirmatively that the testator knew and approved of the contents of the document, and it is only where this is done that the onus is thrown on those who oppose the will to prove fraud or undue influence, or whatever else they rely on to displace the case made for proving the will.”

214.  An appreciation of the facts of that case is important.  The testatrix’s cousin sought to propound what was unquestionably her will.  Two defendants, including one J Painton, sought to propound a will made two days later which was said to have revoked the former will, giving the bulk of the estate to J Painton and making one of his sons sole executor.  This latter will was in the handwriting of Thomas Painton, another of J Painton’s sons.  It was he who had persuaded the testatrix to sign it and he who, along with a friend of his, purported to witness the will. 

215.  Thomas did not take anything under the will, having been an attesting witness, but his father took nearly all the property it disposed of.  It was in this context that Lindley LJ stated that the relevant rule in Barry v Butlin was not confined to cases where the person who obtained the will was also the person who benefited under it.

216.  Accordingly, what the English Court of Appeal decided inTyrrell’s case was that where evidence (not narrowly confined to evidence of direct benefit to the very person obtaining the will) existed raising the issue whether the testatrix knew and approved the will’s contents, the person propounding the will had to adduce affirmative evidence of such knowledge and approval sufficient to “remove the suspicion” and therefore to discharge his persuasive burden of proving the will.  The judge was criticised for treating the sole issue as one of fraud, upon which the testatrix’s cousin bore the burden, without considering whether J Painton had discharged his prior burden of establishing the requisite knowledge and approval on the testatrix’s part.

(a) Thus, the context of the passage in Lindley LJ’s judgment cited above was as follows.  His Lordship stated:

“The circumstances under which it was executed are so strange and suspicious that I am in the greatest doubt whether she knew the effect of the document she was signing.” (at 156)

And after considering the evidence, he asked rhetorically:

“Can any one doubt that Mrs. Bye did not know what she was doing when she executed this said will?” (at 156)

His criticism of the judge for failing to call for affirmative evidence of knowledge and approval was put as follows:

“He seems to have thought that he had only to address himself to the question of fraud - whether the will had been obtained by fraud; that the burden of proving that lay on the plaintiff, and that he had not discharged himself of that burden. Now, in my opinion, this will of the 9th was executed under such suspicious circumstances that he ought to have said, ‘Do the defendants affirmatively establish to my satisfaction that the testatrix knew what she was doing when she executed this will?’” (at 156)

(b) A L Smith LJ similarly noted that the judge had not dealt with the prior issue, commenting that the positive evidence of knowledge and approval – given by Thomas Painton and his friend – were nowhere sufficient to enable his father to discharge his burden (at 158).  Davey LJ was in complete agreement (at 160). 

217.  This view of the confined effect of Tyrrell’s case corresponds to the view taken by Scarman J in In the Estate of Fuld, decd (No 3) [1968] P 675 and in Fuller v Strum cited above.  In Re Fuld, for the purposes of determining a conflicts of law point, Scarman J held that the rule in Barry v Butlin is evidential in character, which follows from the fact that it displaces the usual presumption of knowledge and approval and activates the burden on the proponent of the will affirmatively to prove such knowledge and approval (at 698).  The Tyrrell case was cited in the context of counsel seeking to distinguish Barry v Butlin on the ground that there was no evidence that his client, a Dr Tarnesby, had drawn the relevant codicil which named him as a beneficiary.  Scarman J pointed to Tyrrell as showing that the rule was not confined to cases involving direct benefit to the very person who had drawn up or obtained the will (at 712).

218.  It follows that properly understood, Tyrrell extends the second rule only to the extent of embracing the situation where the person who had drawn up the will was someone other than the person who took under it, but where the two were closely related or associated with each other.  In such cases the proponent of the will was equally required to dispel that suspicious circumstance by adducing affirmative evidence that the testator did indeed know and approve the contents of the will.  The case is not authority for any broader rule, far less authority for requiring the person propounding the will to dispel any circumstances which might be said to be suspicious in relation to the will propounded.

219.  In the respondent’s printed case, however, it was submitted that Tyrrell is authority for the “well-established principle that whenever a will is prepared in circumstances raising a suspicion that it does not express the mind of the testator, the Court ought not to pronounce in favour of it unless that suspicion is removed.”   It relies in support on the decision of the Victorian Court of Appeal in McKinnon v Voigt [1998] 3 VR 543,  where Tadgell JA stated:

“The case of Tyrrell v Painton illustrates very well the point that, where there is cast upon a propounded will a well-founded suspicion that it might not represent the mind and will of the person whose will it is represented to be, no grant will be made before the circumstances which raise the suspicion are explained so as to dispel it. The principle is the same whatever be the facts and circumstances which create the suspicion. It must be so whether the objection is, on the one hand, that the document, although executed by the deceased person, does not for some reason represent his mind and will or whether it is, on the other hand, that the document was not executed by the deceased at all.”  (at 556) 

220.  In consequence, according to Tadgell JA, where forgery is alleged, a two-stage process must be undertaken.  His Honour stated:

“In the present case the question whether the disputed will was proved by the appellant to be a forgery did not arise for the judge’s consideration unless and until the suspicious circumstances had been explained and the suspicions dispelled.” (at 557)

221.  With respect, such a broad view of the second rule is not justified upon a proper understanding of Tyrrell and I am unable to accept the correctness of McKinnon v Voigt in the present context.  That case was adopted as authority for a similar two-stage approach urged in the respondent’s printed case, namely, that unless and until the appellant had dispelled all suspicious circumstances surrounding the will, there was no burden to show forgery resting on the respondent.  However, at the hearing before this Court, Mr Thomas did not press that argument, acknowledging that “analytically that may have been inappropriate”.  He invited the Court to conclude that notwithstanding any errors of approach there may have been, the findings of Yam J and the Court of Appeal were wholly justified.  His submission was as follows:

“But the important thing, the only important thing is this: that we also say that the Court of Appeal would have reached exactly the same conclusion if the position had been analysed differently.  The strength of the evidence we say was such that however you look at it, they would have found that this will was a forgery.”

222.  It was Mr Thomas’s submission that whatever references may also have been made to “dispelling suspicious circumstances”, both Yam J and the majority in the Court of Appeal had effectively adopted the Re H standard in reaching their respective decisions that the 1990 Will was a forgery, an approach which was unimpeachable.

H.      The approach of the courts below to the evidence

H.1    Yam J’s approach to the factual evidence 

223.  I am unable to accept Mr Thomas’s submission.  It is true that Yam J stated (at §6.14) that Re H is applicable and that the court “should require strong and convincing evidence before it would make a finding of forgery”.  However, that was said in relation to the second half of his judgment concerning the expert handwriting evidence, to which I will later return.  In relation to his analysis of the factual evidence, there can be no doubt that Yam J’s approach was erroneously to impose a burden of dispelling suspicious circumstances on the appellant.

224.  In his introductory chapter (at §1.14), Yam J described the main issues in the following terms:

“As regards the 1990 Will, the main issues are :

(a) whether the signatures of the deceased on the 1990 documents were signed by the deceased;

(b) whether the signatures of the witness Mr Tse were genuine. This issue would not per se affect the validity of Documents A, B or C as a Will, but it would have some important bearings on the previous issue (a); and

(c) whether the defendant could dispel all the suspicious circumstances surrounding the 1990 Will, including its execution and surfacing, and her subsequent conduct in propounding the 1990 Will.”  (Italics supplied)

225.  It seems clear that Yam J considered the imposition on the appellant of a burden of dispelling suspicious circumstances to be mandated by authorities in the Barry v Butlin line of cases.  There had apparently been “severe, lengthy and voluminous arguments” on “suspicious circumstances” from both sides concerned with the pleadings, with the relevance of knowledge and approval and the burden of proof (§5.1).  However, regrettably, Yam J brushed aside such arguments as unnecessary, stating:

“I consider that all these arguments were unnecessary in this case including the argument on burden of proof. The case before me is simply this : The defendant in her counterclaim relied on the 1990 Documents as the last will of Wang. The plaintiff on the contrary pleaded that these documents were nothing but forged documents. In other words it is not a case of just lack of knowledge and approval of the testator in signing the documents which happened to be a testamentary document. It is an outright attack on the Documents that they were not the testator’s testamentary documents at all and ex hypothesi the testator certainly would have no knowledge and approval of the Documents in question. Thus all these suspicious circumstances, to my mind, were raised to support the plaintiff’s contention that the 1990 Documents were all forged.” (§5.2 – Italics supplied)

226.  Yam J cited Fuller v Strum [2002] 1 WLR 1097 and Re Fuld, Decd  (No 3) [1968] P 675, which have both been discussed above.  After quoting various passages from those judgments, but without any analysis, he stated:

“In light of the aforesaid two cases, and applying those legal principles as therein decided, the first question in this case is :

What are the suspicious circumstances which I find to be suspicious basing on those facts I found to be true.

The second question is therefore :

With all these suspicious circumstances whether I am satisfied that the defendant has dispelled all these suspicious circumstances hovering over the 1990 Documents before they were admitted to probate.

Alternatively, whether those suspicious circumstances would enable me to say that it is highly probable that the 1990 Documents are just forged documents.” (§5.11)

227.  It is impossible to see how the propositions just quoted can be said to flow from anything stated in the two authorities cited.  One can only presume that those propositions derive from a fundamental misunderstanding of the second rule in Barry v Butlin.

228.  In any event, using the questions which he posed as his framework for approaching the factual evidence, Yam J identified nine “Suspicious Circumstances” which he numbered from I to IX, and, after discussing the evidence relating to each of those matters, concluded:

“The defendant by her counterclaim propounded the 1990 Will by these four 1990 Documents. The burden is on her to prove that these were the last testamentary scripts of Wang, the deceased, and that they were genuine documents properly executed by the testator. I am afraid that the evidence before me is far from that. She has failed to dispel all these suspicious circumstances in order to prove the 1990 Will before it could be admitted to probate. Those suspicious circumstances together can only point to one irresistible conclusion and that is the 1990 Documents are nothing but forged documents.” (§5.127, Italics supplied)

229.  The Judge went on to state that the undispelled suspicious circumstances would compel the same conclusion even if the respondent bore the burden of proving forgery:

“If the burden is on the plaintiff to prove the 1990 Documents were forged documents as he has alleged forgery in his pleadings, I must say that he has discharged this burden by the evidence presented before me. All these pieces of evidence concerning suspicious circumstances alone are strong and cogent evidence.  The only irresistible conclusion that one can draw from all the unanswered or unanswerable, or unexplained or unexplainable questions raised by the evidence presented by the plaintiff, is that the 1990 Documents are forged documents. In order words, even assuming that the plaintiff has no evidence whatsoever on handwriting to prove that the signatures were not written by Wang nor Tse respectively, all the suspicious circumstances raised by the plaintiff would by themselves sufficient to cast grave doubts on the authenticity of the 1990 Documents......” (§5.128 – Italics supplied)

230.  The logic of that passage is hard to follow.  It is difficult to see how, if the respondent bore the burden of proving forgery, he could discharge it by pointing to “unanswered or unanswerable, or unexplained or unexplainable questions”.  In reality, it represents simply another instance of the Judge making findings on the basis of the appellant’s failure to dispel suspicious circumstances.

231.  A feature of Yam J’s judgment involves his tendency to terminate a discussion of factual evidence with a series of questions posed by him but left unanswered in relation to matters which he had obviously identified as “suspicious”.  This, it appears, was treated as damning in relation to the appellant’s case, she having failed to dispel those suspicions.

232.  This approach was consistently applied in relation to the facts.  At the conclusion of his judgment, Yam J states:

“31.33    The aforesaid suspicious circumstances are so compelling. The only irresistible conclusion that one can draw from all the unanswered or unanswerable, or unexplained or unexplainable questions raised by the evidence presented by the plaintiff, is that the 1990 Documents are forged documents. In other words these suspicious circumstances would by themselves sufficient to cast grave doubts on the authenticity of the 1990 Documents without any other evidence in handwriting.

31.34    In the end, I find that the defendant has failed to dispel all these suspicious circumstances in order to prove the 1990 Will to be admitted to probate.”

233.  It goes virtually without saying that the imposition of such a burden on the appellant fatally flaws Yam J’s treatment of the factual evidence.  As discussed above, the appellant, as the person propounding the 1990 Will, bore the burden of proving its execution by Mr Wang on the balance of probabilities.  She had to show, at the end of the day, that it was more likely than not that it was his signature.  That is wholly different from seeking to identify “suspicious circumstances” and then requiring her to dispel them.  The evidential burden on the respondent was to adduce evidence which, properly assessed, was capable of justifying a reasonable inference that the relevant signatures were forged.  It was not enough merely to raise “suspicious circumstances” about matters surrounding the contested will.  That is an approach which places a burden on the proponent which is potentially impossible to dispel, human ingenuity being such that, with little effort, a procession of supposedly suspicious circumstances and unanswered questions can be paraded before the court.  Such a burden is arguably more stringent than the criminal standard requiring proof beyond reasonable doubt.

234.  This is particularly so in a case like the present, where the challenge to the genuineness of the signature is based solely upon inferences drawn from circumstantial evidence.  The Lee Ming Tee principle emphasises the unacceptability of drawing inferences of serious misconduct such as forgery on the basis merely of suspicion, an approach which immediately lends itself to conjecture and prejudice. 

235.  It may be noted that in Worth v Clasohm (1952) 86 CLR 439, a probate case where the testamentary capacity of the testatrix was in issue, the Australian High Court pointed to the error of requiring the propounder of the will to dispel all residual doubts, equating such a burden with the criminal standard of proof:

“A doubt being raised as to the existence of testamentary capacity at the relevant time, there undoubtedly rested upon the plaintiff the burden of satisfying the conscience of the court that the testatrix retained her mental powers to the requisite extent.  But that is not to say that he was required to answer the doubt by proof to the point of complete demonstration, or by proof beyond reasonable doubt.  The criminal standard of proof has no place in the trial of an issue as to testamentary capacity in a probate action.  The effect of a doubt initially is to require a vigilant examination of the whole of the evidence which the parties place before the court; but, that examination having been made, a residual doubt is not enough to defeat the plaintiff’s claim for probate unless it is felt by the court to be substantial enough to preclude a belief that the document propounded is the will of a testatrix who possessed sound mind, memory and understanding at the time of it execution.”  (Per Dixon CJ, Webb and Kitto JJ at 453)

H.2    Yeung JA’s approach to the factual evidence 

236.  While Yeung JA’s approach to the factual evidence was more complex, it is clear that he gave his support to the criticised approach adopted by Yam J. 

237.  Yeung JA correctly stated that the burden of proving execution by Mr Wang rested on the appellant, to be discharged on the balance of probabilities (§4).  He also took the view that the issue of genuineness and that of forgery were two different issues, the respondent bearing the burden of proving the latter on a Re H standard (§7):

“If ‘the 1990 documents’ were forged, obviously they could not have been duly executed, although the reverse was not necessary true as technically it involved a different burden and/or standard of proof.” (§8)

238.  His Lordship noted Yam J’s reliance on the appellant’s failure to dispel suspicious circumstances as the basis for his concluding that the documents were forged (§244).  Acknowledging and partially accepting Mr Michael Thomas’s criticism of Yam J’s imposition of such a burden on the appellant (§249), he held that Yam J should not have based his finding of forgery on such suspicious circumstances alone (§409). 

239.  However, equating Yam J’s items of “suspicious circumstances” with what Yeung JA called “opposing circumstantial evidence” (meaning circumstantial evidence adverse to a finding of genuineness (§320, §333 and §347), he held that such suspicious circumstances were nevertheless highly relevant to the question of due execution (§§324-325). 

240.  Yeung JA therefore drew a distinction between making a finding of forgery based merely on the “suspicious circumstances” (which he rejected) and making a finding that the appellant had failed to prove due execution on the basis of such undispelled suspicions (which he accepted):

“As the propounder of ‘the 1990 documents,’ the Wife had to prove, on a balance of probability, that Wang had duly executed those documents as his last will.

In order to successfully discharge such onus, the Wife might have to dispel the ‘suspicious circumstances’ if such ‘suspicious circumstances’ were inconsistent with the due execution of ‘the 1990 documents’. The Wife might have to prove that Wang’s questioned signatures in ‘the 1990 documents’ were genuine as part of her duty to establish due execution satisfactorily.

However, the Wife had no burden to disprove the allegation of forgery based on the ‘suspicious circumstances,’ even if they were expressly pleaded (which they were not). In fact, it was the Father who had the burden to establish forgery with strong and cogent evidence.

The distinction between the aforementioned two points (i.e. the burden to dispel the ‘suspicious circumstances’ to prove due execution and the burden to disprove the allegation of forgery based on such ‘suspicious circumstances’) may be a fine one but should not be overlooked.” (§§414-417)

He added:

“Although I rule that the judge was not entitled to find forgery based solely on the ‘suspicious circumstances,’ this does not of course mean that the ‘suspicious circumstances’ identified by the judge were irrelevant.

The judge found ‘the 1990 documents’ to be forged based on the ‘suspicious circumstances’. A fortiori it must be his conclusion that the ‘suspicious circumstances’ were inconsistent with the Wife’s case of due execution of ‘the 1990 documents’ by Wang.  (§§423-424)

He concluded :

“In my view, the judge had valid bases to conclude that the circumstantial evidence identified by him did not support the due execution of ‘the 1990 documents’.” (§487)

241.  Yam J’s approach was therefore resuscitated on the basis that while his finding of forgery on the basis of the suspicious circumstances could not be accepted, the appellant’s failure to dispel the itemised suspicions justified the conclusion that she had failed to discharge her burden of proof.  Yeung JA’s approach was therefore also seriously flawed by his imposition in this way of a burden “to dispel suspicious circumstances” on the appellant. 

H.3    Yuen JA’s approach to the factual evidence 

242.  Yuen JA correctly attributed the legal or persuasive burden of proving the will on the balance of probabilities to the appellant (§872) and an evidential burden to the respondent of establishing his allegations of forgery on a Re H standard (§§871-872).  She also correctly held that the Barry v Butlin line of authority was concerned only with the requirement that the person propounding the will should prove that the testator had the requisite knowledge and approval of the will’s contents and that this element of proof was not in issue in the present case (§§889-893). 

243.  However, she fell into error in asserting the relevance of “suspicious circumstances” in the following terms:

“That [ie, the irrelevance of the knowledge and approval issue] being the case, the body of case law relating to ‘suspicious circumstances’ in the context of want of knowledge and approval is not, in my view, relevant.

However, the body of evidence which might be termed ‘suspicious circumstances’ in a ‘want of knowledge and approval’ case may still be relevant.  I do not agree with the trial judge that it was for the Wife to dispel the ‘suspicious circumstances’ such that failure to do so would prove forgery.  But the body of evidence is still relevant to the issue of due execution as ‘all collateral circumstances’ have to be considered (Saph v Atkinson (1822) 1 Add 162, 163).  The question is: tested against that evidence, has the Wife discharged the burden of proving that the Deceased had signed the propounded documents as his will?” (§§894-895)

244.  This must be understood in the light of her Ladyship’s earlier formulation of how a decision should be reached on the evidence, where she stated:

“If the Father succeeds in proving that the signature of the Deceased has been forged, then obviously the will has not been proved.  However, even if the evidence is not strong enough for the Father to prove forgery, the evidence he has adduced towards that end, together with other evidence, may still instill such doubts in the court’s mind that the court finds that it cannot grant probate of the will being propounded.” (§874)

245.  Yuen JA’s approach was therefore effectively the same as that of Yeung JA.  She rejected Yam J’s finding that the appellant’s “failure to dispel suspicious circumstances” was sufficient to ground a finding of forgery but held nevertheless that “the body of evidence which might be termed ‘suspicious circumstances’” was material to the decision whether the appellant had discharged her burden of proving due execution: such undispelled suspicious circumstances may “instill such doubts in the court’s mind” that it refuses the grant of probate.  Instead of asking whether, taking the evidence as a whole, assessed in accordance with the Re H and the Lee Ming Tee principles, it is more probable than not that the signature is genuine (whatever residual doubts there may be), Yuen JA’s approach involves asking whether the “body of evidence” involving “suspicious circumstances” instills doubt in the court’s mind.  This wrongly treats unanswered questions as a valid basis for refusing relief. 

H.4    The approach to reviewability on appeal 

246.  It follows from the foregoing discussion that the approach to the facts adopted by both Yam J and the majority in the Court of Appeal was tainted by the misapplied notion of “dispelling suspicious circumstances” so that their findings are subject to appellate review. 

247.  The present case turns crucially on the determination of what inferences can properly be drawn from proven facts and on the evaluation of expert evidence.  In relation to certain issues, criticism has been levelled at the Judge for making findings without a proper evidential basis or for drawing inferences unjustified by the primary facts.  These complaints do not relate to assessing the credibility of any witness testifying in the witness-box (with the minor exception of Dr Kung, mentioned below).  Accordingly, it was correctly accepted that the Court of Appeal was in as good a position as the trial judge to evaluate the evidence as a whole (§§536 and 869).  It was properly able to substitute its own findings if satisfied that Yam J’s findings were reached on the wrong legal footing, or without proper evidence or involved unjustified inferences or were otherwise not supportable.  The same applies to this Court in disposing of the current appeal.

248.  The relevant principles are stated in a well-known passage from Lord Reid’s speech in Benmax v Austin Motor Co Ltd [1955] AC 370 at 376:

“...... in cases where there is no question of the credibility or reliability of any witness, and in cases where the point in dispute is the proper inference to be drawn from proved facts, an appeal court is generally in as good a position to evaluate the evidence as the trial judge, and ought not to shrink from that task, though it ought, of course, to give weight to his opinion.  In Riekmann v. Thierry (1896) 14 RPC 105, 116,  Lord Halsbury said: ‘The hearing upon appeal is a rehearing, and I do not think there is any presumption that the judgment in the court below is right.’ And later in the same speech he said: ‘Upon appeal from a judge where both fact and law are open to appeal, it seems to me that the appellate tribunal is bound to pronounce such judgment as in their view ought to have been pronounced in the court from which the appeal proceeds, and that it is not within their competence to say that they would have given a different judgment if they had been the judge of first instance, but that because he has pronounced a different judgment they will adhere to his decision.’”

249.  In reviewing the findings of Yam J and the majority in the light of the evidence, it is important to bear in mind the ultimate factual issue in this case: did Mr Wang sign the 1990 Will?  A discriminating approach must be adopted to the copious evidence adduced. 

(a) Plainly, the most important class of evidence is that which bears directly on that issue.  This consists of the evidence of Mr Tse as an attesting witness and of the appellant concerning the provenance of the will. 

(b) The Court must examine the admissibility, nature and content of the materials relied on and the circumstances in which such direct evidence came into being, assessed in the light of attacks mounted by the respondent who contends that the probabilities, both inherent and based on the circumstantial evidence, militate strongly against accepting such evidence. 

(c) A great deal of the evidence canvassed at trial is much more remote.  This involves evidence adduced in support of the rival contentions that making a will like the 1990 Will was – or was not – the sort of thing that Mr Wang would be likely to have done.  Except in quite exceptional cases, such evidence is likely to be of much less weight and importance and is likely to be unable to override definite evidence in the direct categories.

(d) Finally, there is the category of expert opinion evidence as to whether, on the basis of examining the documents in question, the signature of the 1990 Will was – or was not – that of Mr Wang.  As with evidence about the testator’s character and disposition, such opinion evidence, while obviously meriting respect, is inherently of less weight and importance than the direct evidence, particularly where the experts are divided and their evidence inconclusive. 

I.       The evidence relating to Mr Tse as attesting witness 

I.1     The facts

250.  Mr Tse had joined Chinachem as an employee between 1972 and 1974, and had returned to that company in 1982.  From 1983 onwards, after the first kidnapping, while formally employed by Chinachem, he worked as a personal manservant to the Wang couple, residing in their home.  He was constantly at their side.  He accompanied them to and from work each day, cooked and ran personal errands for them.  When Mr Wang was hospitalized after his riding accident on 10 March 1990, Mr Tse was at his bedside.  After Mr Wang’s kidnapping on 10 April 1990, he continued to work in a similar capacity for the appellant.

251.  Mr Tse was married to Madam Tong Kwai King (“Mrs Tse”) and in 1992, she became the registered shareholder of one of two subscriber shares in a company called Global Cheer Limited.  There is not now any dispute that she in fact held that share as a nominee for the appellant’s sister-in-law and that she ceased to be a shareholder in 1993.

252.  When, on 6 September 1999, the sealed envelope was opened in court by Yam J and Mr Tse’s signatures were seen on the 1990 Will,  Joseph Poon arranged for copies of the 1990 documents to be taken and sent a set of them to the appellant.  He then showed copies to Alex Lam, a solicitor in his firm, and asked him to go to Chinachem to take a statement from Mr Tse as to the circumstances of his signing the will.  Alex Lam was not involved in the litigation and had no familiarity with the issues.  He testified that he was not given any copies to take with him since he was told that the appellant would have copies.

253.  Alex Lam arrived at Chinachem that evening after the offices had closed.  He was led to a large conference room where he saw “an old gentleman” sitting.  It was Mr Tse.  The appellant joined them shortly afterwards and handed them copies of the 1990 documents, telling Mr Tse that Alex Lam was a solicitor who would have questions to ask him.  She then left the conference room.  Alex Lam showed Mr Tse each of the 1990 documents in turn, and Mr Tse identified his signatures on each of them, circling the signature and writing in Chinese in the margin of the document each case: “This is my signature”, signing under that statement and drawing an arrow to indicate the copy signature in question.

254.  Alex Lam testified that he asked Mr Tse four questions which were “roughly (i) when he worked at Chinachem and (ii) what his responsibilities were, (iii) how these documents came about, and (iv) how his signatures came about” (the numbering has been inserted for convenience).  Mr Tse gave oral answers and, when asked whether he wished Alex Lam to write them down, Mr Tse chose to write out his statement by himself.  This took about an hour without any interruption by or help from Alex Lam.  The statement was then put into the form of a statutory declaration which, contrary to best practice prescribed by the Law Society, Alex Lam witnessed as solicitor.

255.  Mr Tse’s statutory declaration, written in his own handwriting in Chinese, neatly and with virtually no mistakes, reads in translation as follows (with paragraph numbering inserted for convenience):

“1.    I, Tse Ping Yim, male, aged 69, HKID No XXXXXXX(X), have been working in the household of Mr and Mrs Wang Teh Huei as a butler since 1983 and thus I have to live in the house of Mr Wang on a long-term basis.

2.    At the early stage, in about 1972 or 1973, I worked in the construction sites of Chinachem and then left Chinachem for some period of time.  In 1982 I went back to work in Chinachem again.  I have been working in the household of Mr Wang Teh Huei for quite a long period of time and feel that Mr and Mrs Wang treated me like their family members.  I always saw them going out and coming back together and walking hand in hand.  They were an affectionate couple.

3.    In March 1990, Mr Wang fell from a horse.  He was first admitted to Queen Mary Hospital at about 8:00 in the evening and then transferred to St Paul’s Hospital for treatment by neurosurgical specialists.  On that evening up until 3 am, I was with him in the hospital all along.

4.    He remained in the hospital for about 3 or 4 days before (he) was discharged.

5.    Having been discharged from the hospital, Mr Wang Teh Huei immediately returned to work in his office as usual.  I accompanied Mr and Mrs Wang to go to and from the office as usual day in, day out.  Every day I would dispatch documents and deposit cheques for the company if I was free.  I vaguely remembered that after Mr Wang Teh Huei was discharged from hospital, Mrs Wang told him not to do exercise for the time being because his wound had not healed yet.

6.    During that period of time, Mr Wang Teh Huei once asked me to go into the glass partitioned office where he worked.  At that time there were only two persons, Mr Wang Teh Huei and me, inside the room.  Mr Wang Teh Huei asked me to sign several documents already with words written thereon.  I did not read the contents of the documents because I absolutely trusted Mr Wang Teh Huei.  And, Mr Wang Teh Huei did not explain the contents thereof.  I saw Mr Wang Teh Huei signing the documents and I signed the documents too at the position designated by Mr Wang Teh Huei.  The signed documents were left on the desk of Mr Wang Teh Huei and I immediately left the office of Mr Wang Teh Huei.

7.    On September 6, 1999 when I was about to get off work, lawyer Lam Fong Vai showed me four documents, namely Appendix 1, Appendix 2, Appendix 3 and Appendix 4, with the signatures of Tse Ping Yim thereon.  Having seen (the signatures), I confirm that all of which were affixed by me.

8:30 pm on September 6, 1999.”

256.  On 9 September 1999, it was decided on leading counsel’s advice that a statement should also be taken from Mr Tse by Lovells who then had conduct of the swear death proceedings.  It was taken by Allan Leung, a partner in the firm, with the assistance of a solicitor called David Chu.  Allan Leung interviewed Mr Tse and testified that they had a normal conversation without any apparent nervousness or hesitation on Mr Tse’s part, with Allan Leung asking open, and not leading, questions.  This time the answers were recorded by the solicitors in Chinese and Mr Tse later made an affirmation confirming his exhibited statement. 

257.  In his statement, Mr Tse referred to and confirmed the statement that he had given to Alex Lam and, when shown copies of the 1990 documents by Allan Leung, again identified the signatures they bore as his own.  The material paragraphs of his affirmation read (in translation) as follows:

“5.    It was March 1990.  One day in the office at 6:00 or thereabout ie, the time for going off work, Mr Wang Teh Huei asked me to go into his office, a glass partitioned room.  When I got in, I saw a pile of documents on the desk, which was a small pile.  He asked me to sign those documents.  I remember that he signed each document and then asked me to countersign. 

So I followed his instruction and signed the documents.  At that time I did not ask him anything nor did I try to see what the documents were or what were written thereon because I absolutely trusted him.  All along he had been treating me very well, like family members.  He was then in a very bad mood.  He could not do any exercise because of his fall from the horse.  After I had signed the documents per his request, I put them back on the desk and then left.

6.    That was the first time Mr Wang asked me to sign documents although I very often had to sign some receipts.  I remember that when I signed those documents, there were only Mr Wang and me inside his room.

7.    From the time when solicitor Lam Fong Vai showed me those documents which were the same as Appendixes A, B, C & D on Monday evening of September 6 till I signed this statement, I did not tell others about this matter, except on the occasion when Mrs Wang and solicitor Lam asked me on Monday evening whether or not the signatures on the documents the same as Appendixes A, B, C & D were affixed by me.  I told them that the signatures were affixed by me.”

258.  While the application to swear death was in progress, Mr Tse went on paid leave.  He travelled to the mainland on 14 September 1999 and returned to Hong Kong four days later on 18 September.  In his affirmation, Mr Tse described himself as being “of Zhaoqing, Guangdong” and there was evidence that he was in the process of building a house on the mainland.

259.  On 22 September 1999, he obtained a visa to go to Thailand.  On the same day, the respondent was granted leave to swear to Mr Wang’s death.

260.  On the following day, 23 September 1999, in an application which was brought ex parte on notice to the appellant (and therefore without the prior issue of a summons), the respondent sought and obtained Yam J’s leave to report a case of suspected forgery to the police.

261.  On the next day, 24 September 1999, Mr Tse went again to the mainland.  That was also when the respondent commenced the present probate proceedings.  No forgery report was made to the police by the respondent until some time in October 1999.

262.  On 28 October, Mr Tse went from the mainland to Thailand on a sightseeing tour, returning to the mainland on 2 November 1999.

263.  On 14 November, he consulted doctors in the Shenzhen People’s Hospital and was immediately admitted as an in-patient.  A week later, on 21 November 1999, he was diagnosed to have an advanced case of liver cancer.  He returned to Hong Kong on 4 December 1999 and died two days later.  The police had sought to interview him on his return but his medical condition prevented this from taking place.

I.2     The respondent’s case on Mr Tse 

264.  It is not in dispute that Mr Tse’s statutory declaration was admissible evidence of execution of the will by Mr Wang.  Section 15 of the Oaths  and Declarations Ordinance, Cap 11 provides:

“An attesting witness to the execution of a will, deed or document, and any other competent person, may verify and prove the signing, sealing, publication or delivery of the will, deed or document by a declaration made in the manner provided by section 14”.

And Mr Tse’s affirmation was admitted as hearsay under section 47 of the Evidence Ordinance, Cap 8.

265.  It is obvious that if Mr Tse’s statements are not undermined, they would be of the very first importance as direct evidence of execution by Mr Wang.

266.  As previously noted, the respondent alleged that Mr Tse’s signatures on the 1990 documents were forged.  That allegation was introduced by amendment to his pleadings on 15 November 2000, some 11 months after Mr Tse’s death.  It is unlikely that at that stage the respondent knew that Mr Tse had given statements confirming their authenticity.  Plainly, those statements placed the respondent in a somewhat awkward position.  How could it be maintained that the signatures were forged when Mr Tse had himself confirmed their authenticity?   The suggestion that the signatures were forged but that Mr Tse had nevertheless confirmed that they were genuine implies the existence of a multi-party conspiracy with numerous inherent improbabilities, as discussed in Section G.6 above.

267.  Nevertheless, the respondent persisted in alleging that the Tse signatures were forged throughout the trial.  His three experts all testified in terms of great certainty that those signatures had indeed been forged by superimposition tracing.  Despite the inherent difficulties, which, as Waung J points out (§1088) were never addressed, Yam J had no doubt that such forgery had been established.  Indeed, he made a finding that the Tse signatures on the 1990 documents had been traced “according to a model signature written by him in the later years after Wang was kidnapped in 1990” (§16.7), adding:

“......the evidence in support of forgery in respect of Tse’s signatures are very, very cogent and compelling. I can say that I have no doubt that these four signatures are forged signatures.” (§16.8)

268.  However, in the Court of Appeal, only Yeung JA was prepared to uphold that finding.  Yuen JA held that the evidence of forgery in relation to Mr Tse’s signatures was insufficient.  Waung J (in his dissent) held that the evidence showed that the Tse signatures were probably genuine.  As indicated in Section G.6 above, the forgery allegation was jettisoned before this Court, with the respondent endeavouring to replace it with the unpleaded “subtle form of forgery” conspiracy case. 

269.  It is against this background that consideration must be given to the evidence and arguments relied on for undermining the Tse statements and for supporting the existence of a conspiracy to effect the subtle form of forgery.  The Re H standard and the Lee Ming Tee principle must be kept fully in mind. 

270.  The arguments mounted against the veracity of Mr Tse’s statements, all of which found favour with Yam J, may be grouped under nine headings, namely:

(a)    that it was improbable that Mr Wang would choose his manservant, Mr Tse, to witness his will;

(b)    that Mr Tse’s veracity should not be accepted because he had denied knowing that he was witnessing a will;

(c)    that the nearly immaculately written statement given to Alex Lam had not in fact been written by Mr Tse but had been pre-prepared and tailored untruthfully to the needs of the appellant’s case;

(d)    that the statement was too well-written and too mistake-free not to have been coached and it was therefore untruthful;

(e)    that parts of the contents of the statements were inexplicably volunteered and too well-tailored to the needs of the appellant’s case to be believable;

(f)    that Mr Tse’s credibility should not be accepted because of his wife’s shareholding in Global Cheer Ltd;

(g)    that Mr Tse’s credibility should not be accepted because he failed to mention signing the 1990 Will when questioned by police officers investigating Mr Wang’s kidnapping;

(h)    that Mr Tse’s credibility should be rejected since he had obviously fled the jurisdiction as soon as the prospect of a police forgery investigation was raised; and,

(i)    that the absence of physical evidence of impressions of Mr Wang’s signatures on some of the 1990 documents shows that Mr Tse’s account of events on 12 March 1990 is false.

I.3(a) Mr Tse as an improbable choice as attesting witness 

271.  Yam J held found that it was “very unlikely that [Mr Wang] would confide in [Mr Tse] anything if at all and certainly it was more than unlikely that [Mr Wang] would ask him to be a witness to his will” (§5.82).  This was due to Mr Tse’s “talkative nature” – a quality apparently abhorred by Mr Wang – and which “would mean that he would not keep the secret of what he saw or read” (Ibid).  The attribution of such talkativeness to Mr Tse was based on evidence given by Teresa Wang about a conversation she had had with him on their first meeting when she spent a few days with the Wangs back in 1984.

272.  That is hardly evidence which justifies the inference suggested.  Teresa Wang could speak to no more than a brief encounter with Mr Tse six years before the date of the will.  Her evidence should in any event have been treated with great reserve.  As Yeung JA found, Teresa Wang:

“...... was not an impartial witness. Her description of the Wife demonstrated that she harboured substantial dislike, if not outright hatred, for her.” (§654)

273.  Indeed, the inference Yam J drew runs very much against the objective evidence.  Mr Tse had, since 1983, been working as a personal manservant living in the Wang household and was always at hand to render personal services to the Wang couple.  He would hardly have been kept in that position if he was talkative and indiscreet.  On the contrary, he was precisely the kind of faithful retainer who might be asked to attest his employer’s will.

274.  Yam J’s finding is particularly surprising and unsustainable since he was compelled to accept that:

“There is no doubt that he was trusted by Mr and Mrs Wang, otherwise he would not be allowed to work and stay in their house.” (§5.78)

275.  In his conclusions, Yam J dealt with this point as follows:

“...... why should Teddy select Tse to be a witness at all on these important documents involving all his estate and the two most important persons in his life who were near and dear to him (i.e. his father and his wife), when he did not even let them know the content of the documents? Why did Teddy not ask the two mysterious persons who wrote out Documents A, B and C to witness his signatures instead? Why should he further ask a fourth person, apparently in the name of Lee Kin Sang, to witness his signature on Document A and Document A only?” (§31.28)

276.  This is an illustration of the pernicious effects of imposing a burden on the appellant “to dispel suspicious circumstances”.  Yam J evidently felt that if he could pose a series of questions which were not answered, this could be held against the appellant.  Assessment of the evidence by application of Re H and the Lee Ming Tee principles did not occur.

277.  In the Court of Appeal, only Yeung JA gave any credence to this suggested head of improbability, merely asserting that the choice of a manservant was improbable, without elaboration (§547).

I.3(b) Mr Tse’s alleged denial that he knew he was witnessing a will

278.  There are two aspects to this criticism. The first is straightforward.  The two characters (遺囑) meaning “Will” appear prominently on Document A just above where Mr Tse signed his name beside the two characters (証人) meaning “Witness”.  Accordingly, so the criticism runs, for Mr Tse to say in his statement that he did not know he was signing a will is wholly unbelievable since he could not possibly have missed the aforesaid characters.  This is said to cast general doubt on his credibility.

279.  The second point runs like this: Assuming that Mr Tse’s statement is to be believed when it says that he did not know he was signing a will, there could be no conceivable innocent explanation for him choosing to volunteer the information included in his statement.  This is dealt with under point I.3(e) below.

280.  The first criticism depends on the proposition that in his statements, Mr Tse was in fact saying that he did not know he was signing a will and so necessarily implying that he had missed the prominent characters set immediately above his own signature.  Whether this is fair depends on what Mr Tse actually said.

(a) In the statement given to Alex Lam, the material words are:

“Mr Wang Teh Huei asked me to sign several documents already with words written thereon.  I did not read the contents of the documents because I absolutely trusted Mr Wang Teh Huei.  And, Mr Wang Teh Huei did not explain the contents thereof.  I saw Mr Wang Teh Huei signing the documents and I signed the documents too at the position designated by Mr Wang Teh Huei.”

(b) In the statement taken by Allan Leung, the relevant words are:

“So I followed his instruction and signed the documents.  At that time I did not ask him anything nor did I try to see what the documents were or what were written thereon because I absolutely trusted him.” 

281.  I am unable to read into these words the premise of the criticism stated in the preceding paragraph.  Mr Tse was saying that he did not read the documents’ contents because, trusting Mr Wang, he was prepared to sign without doing so.  But he was not thereby suggesting that he did not see characters which could not be missed.  The sense of his words is made clear in the second statement, namely, that he was not trying to see what the documents were, which is an averment that he did not wish to pry or to be inquisitive as to the contents of the documents.  But he was again saying nothing about what he must inevitably have seen.  This point is therefore without substance.

I.3(c) The statement given to Alex Lam as having been pre-prepared 

282.  As noted in Section I.1 above, Alex Lam testified that he asked Mr Tse four questions and then left him to write out his statement on his own.  The resultant statement, in Mr Tse’s own handwriting, aroused Yam J’s suspicions on two bases: first, that it had been written out with hardly any errors in a neat and orderly hand; and secondly, that it dealt with matters which were volunteered by Mr Tse and seemed obviously tailored to support the appellant’s case.  As noted below, some of the Court of Appeal judges had similar comments to make.

283.  However, the inference that Yam J drew from these observations (together with certain expert-type suggestions he made about indentations caused by the signatures in the paper used) was startling: he held that the statement had been pre-prepared and simply produced for signature by Mr Tse (§§5.97-5.98).

284.  To arrive at this conclusion, Yam J had to find that Alex Lam was perjuring himself since he had testified to Mr Tse writing out the statement in his presence.  This Yam J was prepared to do on the basis first, that Alex Lam “gave his evidence in a very soft and inaudible voice” and “appeared to be nervous and timid”; secondly, that he was able to produce to the court neither the copy documents that Joseph Poon was said to have shown to him prior to his interview with Mr Tse, nor any attendance note or bill of costs in relation to that interview; and thirdly, that he had acted as witness to the statutory declaration.  The explanation that the abovementioned documents had gone missing and may have been lost as a result of a burglary at the firm or mislaid when the firm moved its premises was rejected.  Yam J permitted himself to speculate that there might be something in those documents which would “show that the account he gave on what he did in the office of Chinachem was not true” (§5.99).  He concluded:

“......it is extremely doubtful whether the handwritten statement of Tse was produced in the way as described by Mr Alex Lam, the solicitor who took the statement from Tse on 6 September 1999. The evidence suggested that Tse very probably had already prepared his statement written in an entirely neat and orderly manner before this interview.” (31.26)

285.  The evidence suggested nothing of the sort and this was a conclusion reached on an entirely speculative basis.  Moreover, it implicitly postulates a further broadening of the conspiracy theory to include the solicitors Alex Lam and Joseph Poon as participants.  Regrettably, Yeung JA in the Court of Appeal appears to have condoned this approach, demonstrating in the process the perniciousness of a search for “suspicious circumstances” and negativing any application of the Re H principle.  He stated:

“Although the judge might have criticized Mr. Lam harshly, the circumstances in which Mr. Lam took the statement from Tse on 6 September 1999 were indeed highly suspicious......” (§482)

286.  The other members of the Court of Appeal rejected the finding that the statement was pre-prepared.  Yuen JA having cogently demonstrated the lack of substance in the various suspicions voiced (at §§954-961), concluded:

“In my view, evidence of that quality did not justify what was tantamount to a finding that Lam (a professional person with no relationship to any of the parties) had committed perjury.” (§962)

Likewise, Waung J stated:

“......There is absolutely no reason for these two professional men to tell any untruth about what happened on the 6th and 9th September.  It is possible that Alex Lam as a nervous witness might have got some of the details of the giving of the Statement and the making of the Affirmation not totally accurate when he was giving evidence.  But that is not an adequate reason to dismiss the Tse’s Statements and Affirmations.  They were there and could not be ignored and that evidence must be properly weighed.” (§1087)

287.  Mr Thomas did not seek to uphold the finding against Alex Lam before this Court.

I.3(d) The statement as being too well-written and mistake-free to be true

288.  While she dissociated herself from Yam J’s unjustified conclusion that the statement had been pre-prepared, Yuen JA nevertheless expressed some scepticism as to Mr Tse’s ability, without coaching, to produce such a well-written and mistake-free document.  This is echoed in the respondent’s printed case.

289.  The fact that the document was well-composed and lacked mistakes is an odd basis for suspicion, especially if one assumes (as one must) that Alex Lam was not lying and that Mr Tse did indeed write it out himself.  One has to ask rhetorically: so what if he did not make mistakes?  It was common ground between the parties that Mr Tse’s Chinese handwriting was superior to the Chinese handwriting found in the statement written out by a solicitor at Lovells.  A reasonable inference is that drawn by Waung J, namely, that the neatness of the handwriting and the refinement of expression used in his statement indicate “that the writer was a man of some literary skill and also some intelligence” (§1085). 

290.  In the respondent’s printed case, suspicion is sought to be raised on the basis that Mr Tse’s education ended at primary school level so that it was improbable that he could have written so immaculately.  The appellant challenges the assertion that Mr Tse’s education was so limited.  But the short answer, if Alex Lam is to be believed, is simply that, whatever his formal educational level, Mr Tse had plainly written out the document himself and the results speak for themselves.  It is a document exhibiting the cultivated characteristics referred to by Waung J, so that a limited formal education is a background fact of no relevance.

291.  In actuality, while Yuen JA described the statement as “very well-structured and composed” and called it “a remarkable work for an unscripted first draft” implying a degree of scepticism, she never expressly drew any adverse inference based on those comments.  On the contrary, she allows that “the basic structure could be explained by Tse following Lam’s four questions” and also that Mr Tse was “a person who had obviously practised calligraphy”.  Plainly, this provides no basis for inferring participation by Mr Tse in a criminal conspiracy involving the subtle form of forgery.  However, the fact that her Ladyship mentions this point at all suggests that she may erroneously have considered Mr Tse’s well-presented statement to be a suspicious circumstance militating against the grant of probate.

I.3(e) Inexplicably volunteered contents too well-tailored to the appellant’s case to be believable

292.  Both Yeung JA and Yuen JA considered Mr Tse’s statements to contain inexplicably volunteered information which undermined the credibility of those statements.

293.  Yeung JA put this strongly (§§502-503) although he did not make it clear which parts of the statement were thought to be suspicious:

“It is quite inconceivable that Tse could have, by his own volition and without prompting or coaching of any kind, come up with the statements in question, the contents of which were a perfect match with the Wife’s case.

The four questions allegedly put to him by Mr. Lam could hardly have resulted in Tse’s producing the contents of the statement on 6 September 1999 on his own initiative. There are strong internal marks of untruth, unfairness, and partiality in Tse’s statements.”

294.  Yuen JA (§§963-972) considered Mr Tse’s reference to the Wangs as having been an affectionate couple whom he had seen walking hand in hand to be unwarranted by the questions asked and “curiously incongruous” in the context. 

295.  At the hearing, Mr Thomas, who adopted this point, included as part of the suspiciously volunteered material Mr Tse’s reference to the riding accident and Mr Wang’s hospitalization.

296.  This criticism also prays in aid the second facet of the point mentioned in I.3(b) above, namely that the volunteering of such information was particularly sinister since Mr Tse had denied knowing that he was witnessing a will and so could not possibly explain why he considered the matters volunteered to be relevant.  Like the first point dealt with in I.3(b) above, this argument is premised on a reading of the two Tse statements which attribute to him the claim that he did not know he was witnessing a will.  For the reasons given above, that premise is rejected.

297.  Leaving that last point aside, this attack on the veracity of Mr Tse’s statement is based on two propositions: first, that relevant parts of his statement (particularly that given to Alex Lam) are volunteered and happen to suit the appellant’s case very well; and secondly, that there is no plausible innocent explanation for Mr Tse choosing to volunteer the same. 

298.  I am unable to accept those propositions.  It is a matter of everyday courtroom experience that witnesses often have their own concept of what constitutes a relevant answer, a concept which may differ from that of the cross-examiner or the judge.  A witness may consider himself answering the question put whereas the questioner may consider him to be volunteering irrelevant material. 

299.  In the present case, the questions put by Alex Lam were open questions of a general nature and Mr Tse was left to decide for himself what should be included.  Thus, Mr Tse’s account of Mr Wang’s riding accident and his consequent hospitalization (during which time Mr Tse was at Mr Wang’s bedside) and Mr Wang’s return to the office on being discharged may perfectly reasonably be considered as part of his answer to question (iii) as to “how these documents came about”, rather than information inexplicably volunteered.  If the will is genuine, Mr Wang arranged for it to be signed and witnessed by Mr Tse on the day he was discharged from hospital.  It is perfectly reasonable for Mr Tse to have regarded the accident, hospitalization and the signing of the document in the office in the evening of the day of his discharge as a related series of incidents explaining how the 1990 documents came about.

300.  I would, however, accept that his reference to the Wang couple’s displays of affection was something “volunteered” and going beyond the questions asked.  But I cannot accept that this undermines the credibility of the statement or justifies the inference that Mr Tse was inventing his account of Mr Wang’s execution of the 1990 Will. 

301.  In the first place, the fact that the Wang couple were extremely close and openly affectionate, spending most of their time together, often to be seen walking hand and hand, was entirely borne out by the evidence of witnesses including Donald Cheung, Anthony Cheung and Cecil Chao.  So this was not something Mr Tse was making up.

302.  Secondly, given that his work involved him closely attending to the appellant, it would be wholly surprising if Mr Tse was not aware of the bitter litigation that had broken out between the parties.  It is a reasonable inference that by 6 September 1999, Mr Tse must have known that the respondent was contesting the genuineness of the 1990 Will and that the respondent was making adverse allegations against the appellant and about the relationship between herself and Mr Wang.

(a) Thus, in the abortive probate proceedings, the respondent had filed a Statement of Claim putting it in issue whether Mr Wang and the appellant were in fact validly married (Transcript 13.60.22 – 64.5), this having attracted an indignant refutation from the appellant.  It seems entirely likely that such allegation and the indignation it evoked would have come to the knowledge of Mr Tse.

(b) On 26 March 1998, well before the envelope was opened, the respondent had filed affirmation evidence alleging that the appellant’s claim to have a 1990 will made by Mr Wang was false and that the appellant had “in other words, probably committed a crime”. 

(c) As Yuen JA accepted (§970) “it may be assumed that Tse was aware of litigation between the Wife and the Father from newspaper reports”.

(d) In one extensive article on the dispute published in the 22 January 1999 issue of the widely-read Next Magazine, the respondent is reported as saying that he did not believe the appellant’s claim to have a will in a sealed envelope given to her in 1990.  It would be most surprising if this article did not come to Mr Tse’s attention.

303.  Knowing of the issues, Mr Tse could not have viewed the dispute wholly dispassionately.  As attesting witness his integrity was also under attack.  Unless one presupposes his participation in a conspiracy, his natural reaction, upon being asked to give a statement about his signatures on the will would naturally be to refute the doubts.  Having seen the expressions of affection in the 1990 documents themselves (in Document B: “I love my wife.  She is my dearest in the world”; in Document D: “One life one love”) it is unsurprising that Mr Tse should think it appropriate to corroborate the existence of such affection and to be supportive of the appellant’s position.  It does not follow that his statement was a pack of lies. 

I.3(f)  Mrs Tse’s share in Global Cheer Limited 

304.  There is nothing in this point.  Mrs Tse did not hold the share beneficially but only as nominee for the appellant’s sister-in-law (the wife of Dr Kung).  She had ceased to hold it six years before Mr Tse provided his statements to the solicitors.  So it is wholly fanciful to suggest that there was any connection between the veracity of such statements and that nominee shareholding.  It certainly does not justify inferring criminal conduct on the part of Mr Tse or indeed, by a further widening of the conspiracy theory, inferring complicity on the part of Mrs Tse.

I.3(g) Failing to mention his witnessing documents to officers investigating kidnapping 

305.  Yeung JA thought it significant that Mr Tse:

“...... did not volunteer the information to the police when they interviewed him in May and July 1990, very shortly after Wang’s kidnap and very shortly after Tse allegedly appended his signatures to ‘the 1990 documents’ on Wang’s instruction without knowing their contents.”

306.  For the reasons stated above, it is probable that Mr Tse did know that he was witnessing a will.  However, whether he did or did not know the contents of the documents, there is no basis for regarding the kidnapping as somehow obviously connected to those documents so that a failure to inform the police was suspicious.  If Mr Tse had known that it was a will, its relevance to a kidnapping is objectively hard to see.  Mr Wang would have been assumed to be alive, detained by kidnappers seeking to extract ransom.  His testamentary dispositions were hardly to the point.  On the other hand, if Mr Tse did not know what the documents were, why should he assume they were relevant?  And what could he have told the police about them?  This is another instance of an unfounded suspicion being placed in the balance against Mr Tse’s veracity.

I.3(h) Mr Tse fleeing the jurisdiction 

307.  Yeung JA identified Mr Tse’s departure from Hong Kong on 24 September 1999 as “highly suspicious”, postulating that he must have known that the court had just given leave to report forgery to the police so that his behaviour “gives rise to a compelling inference that he wanted to avoid interview by the police and that those controlling him also wanted to prevent him from such interview.” (§§518-521)

308.  His Lordship comments:

“If Tse did not want to be interviewed by the police and those controlling him did not want him to be interviewed, it is definitely legitimate to ask why.” (§522)

He concludes:

“The judge’s questions on the above suspicions concerning Tse’s movement after 24th September 1999 were certainly legitimate and pertinent to the case, particularly in determining the weight to be attached to Tse’s statements. Hence such questions are highly relevant to the Wife’s case of the due execution of ‘the 1990 documents’.” (§526)

309.  It is to be noted that Yeung JA appears to treat as a fact the existence of persons controlling Mr Tse, although the evidential basis for this is nowhere identified.  He later (at §§597-598) makes it clear that his innuendo is directed at the appellant:

“The evidence clearly suggested that Tse was very much under the Wife’s influence. He was a Chinachem employee and had been the couple’s ‘house servant’ for a long time.

Tse’s sudden departure on 24 September 1999 begged explanation. If it did not occur at the Wife’s instigation, it was unreasonable that she made no enquiries about it. On the other hand, if the Wife had made enquiries, the court should have been provided with the results and/or related information.”

310.  Yuen JA pointed to an absence of explanation for Mr Tse’s stay on the mainland (§976), noting that there was some suggestion that the appellant might have “wished to protect Tse from interference”.  Her Ladyship took the view that by leaving, Mr Tse “made himself unavailable to the Hong Kong police” and that “preventing questioning by the Hong Kong police” undermines the credibility of his statements (§§978-979).

311.  Both of these approaches are unsatisfactory.  Yeung JA’s approach was to identify “suspicious circumstances” or unanswered questions – and not proven fact – and to use them as the basis for drawing inferences of criminal behaviour on the part of both Mr Tse and the appellant.  Yuen JA equated Mr Tse’s departure with his “making himself unavailable” and “preventing questioning”, simply by juxtaposing the date of Yam J’s giving of leave with the date of his departure.  The assumption is that Mr Tse was party to a forgery and had to flee.  Such an inference is devoid of evidential support and is unfair to the deceased.

312.  As Waung J pointed out (§1089), there was no evidence that Mr Tse knew that on the previous day, at an ex parte on notice hearing, the respondent had sought and obtained leave to report forgery to the police.  Nor was there reason to infer that his departure on 24 September was sudden and not previously arranged.  He plainly had prior travel plans.  He had been on leave starting on 14 September 1999 and had already made one visit to the mainland returning to Hong Kong on the 18^th.  On 22 September 1999, he had obtained a Thai visa which he later utilised to go to Thailand from the mainland.  There is no reason to believe that Mr Tse was not intending to return to Hong Kong after his departure on 24 September.  The evidence that he was in the course of constructing a house in Guangdong may have supplied a reason for a relatively long sojourn on the mainland.  No one, including Mr Tse, knew at that stage that he was suffering from cancer or that he would be dead before the year was out. 

313.  It is unfair to suggest that he was preventing others from questioning him.  After the envelope was opened, the respondent’s solicitors could, if they had wished, have sought an interview with him in Hong Kong between the 6^th and the 13^th or between the 18^th and the 23^rd of September.  They did not do so.  The report to the police was in fact not made until some time in October and the police did not seek an interview until Mr Tse was on his deathbed.  If the police had wished to interview Mr Tse urgently, they could have approached the appellant or her solicitors to ask them to arrange for a statement to be taken from Mr Tse, but they never did so.  It is therefore hard to see how one can infer that he was making himself unavailable to be questioned when no one had sought an interview.

314.  Indeed, the inference reasonably drawn from Mr Tse’s willingness to give statements about his witnessing the will and to have them made into a statutory declaration and an affirmation, is that he was willing to testify to the genuineness of the will.  If he had known of a potential police investigation, he would obviously have been aware that those documents must at some stage be made available to the police.  There is no reason to doubt that, if questioned by them, he would simply have confirmed the veracity of those statements. 

315.  Yuen JA’s suggests that Mr Tse might have been afraid to face questioning on his statements because:

“In the hands of a skilled examiner, Tse would have had to explain for instance, how he could claim (and why he claimed) not to have seen the characters for ‘Will’ when he purportedly signed document ‘A’, given that the characters were virtually immediately above his signature (since Tse had not suggested that the Deceased had covered up the contents of the document).” (§978)

However, as previously discussed, that suggestion is based on a reading of Mr Tse’s statements that I have rejected.

I.3(i)  Absence of impressions of Mr Wang’s signatures

316.  The argument runs as follows: 

(a) Documents A, B and C were on very thin paper.  The expert evidence indicated that heavy pen pressure was used to produce the Wang signatures, to the extent that the paper may have been perforated at various points. 

(b) In the statement given to Allan Leung (at §5) Mr Tse states:

“When I got in, I saw a pile of documents on the desk, which was a small pile.  He asked me to sign those documents.  I remember that he signed each document and then asked me to countersign.”

(c) When cross-examined, Allan Leung was asked as to his impression of whether Mr Tse was saying that, on 12 March 1990, when Mr Wang signed the 1990 documents, he had signed each of them when the documents were still arranged in a pile.  Allan Leung agreed that this was the impression he received.

(d) Accordingly, so the argument goes, if that account of events is correct, there would inevitably have been an impression made on documents lower down in the pile of Mr Wang’s signature by the pen pressure exerted when signing on documents at the top of the pile.

(e) Since no such impressions were found, Mr Tse’s account of events is to be rejected.

317.  When taking this point Mr Thomas fairly acknowledged that it rested heavily on Allan Leung’s recollection of his interview with Mr Tse on 9 September 1999.  The evidence is indeed tenuous in the extreme.  Plainly, the statement does not say, one way or the other, whether the documents were still in a pile when Mr Wang signed them.  So Allan Leung was asked on 10 June 2002 to describe the “impression” he had gained on 9 September 1999, 33 months earlier, of what Mr Tse was indicating as to whether the documents were signed in a pile.  It is difficult to see how Allan Leung could be expected to have a specific recollection of such a detail after almost 3 years.  Indeed, it would be surprising if Allan Leung had a recollection which extended beyond anything that had been taken down in the statement itself.  Moreover, for Mr Tse himself to advert to such a detail – which would have been of no apparent significance to anyone prior to being raised by counsel at the trial – he would have had to cast his own mind back to 12 March 1990 some nine-and-a-half years earlier.  Accordingly, in my view, the argument relies on evidence that is simply too insubstantial to support it.  There was simply no acceptable evidence that Mr Wang had signed the documents stacked in a pile rather than singly.

I.4     The status of Mr Tse’s evidence 

318.  My conclusion is therefore that the evidence given by Mr Tse in his statutory declaration and his affirmation has not been undermined by any of the foregoing contentions.  The attacks mounted do not begin to make out a case based on the subtle form of forgery.  Being direct evidence of execution of the 1990 Will by Mr Wang Mr Tse’s statements constitute evidence of crucial importance.

J.       The appellant’s affirmations of testamentary scripts 

J.1     Whether they are evidence in the case

319.  The filing by the appellant of her affirmations of testamentary scripts has been noted.  As she did not testify at the trial, she sought to place reliance on the contents of those affirmations as evidence of the provenance of the 1990 Will.  There was some dispute both in the Court of Appeal and in this Court as to whether those affirmations could properly be treated as evidence of their contents.

320.  Unless the court gives leave for the affidavit or affirmation to be read as evidence in the trial pursuant to O 38 r 2(1), or unless there is agreement between the parties, the affirmation does not take effect at the trial as evidence of its contents.  No application was made to the judge.  However, Mr Vos argues that there was agreement either at the trial or on appeal that those affirmations were admissible although obviously, the weight to be given to them was a matter for the court.

321.  In the course of the trial, those affirmations had been used by the respondent in cross-examination of Dr Kung.  Then, on Day 170, after the evidence had concluded and it had been announced that the appellant was not going to enter the witness-box, there was an inconclusive discussion about the status of her affirmations of testamentary scripts, Yam J having been told that the parties were not agreed as to their status but that written submissions were to be filed.  None were in fact filed.  Plainly no express agreement can be made out of those exchanges.

322.  However, it may be that there was implied agreement that those affirmations could be used for their contents since the respondent used parts of those affirmations as the basis of arguments in written closing submissions inviting the judge to draw inferences adverse to the appellant.  This Yam J did in several places in his judgment, pointing out for instance that she had made no mention of the alleged “holiday wills” in those affirmations (§§5.27-5.29) and noting that “according to her, she had always had the 1990 Will in her possession albeit it was in a sealed envelope” (a matter stated in those affirmations) so that she had no excuse for delays in their filing (§5.115).

323.  Similar use was made of them by the majority in the Court of Appeal (§§473-475 per Yeung JA and §927 per Yuen JA).  However, Yuen JA appears to have taken the view that those affirmations were not in evidence, noting (at §988) that “there had not been any orders from the court that they could be used in place of oral evidence at trial” under the heading “No evidence of transfer of documents from Deceased to Wife”.  Yeung JA, however, recorded a concession by Mr Robert Tang SC then appearing for the respondent, as follows:

“Mr. Tang fairly conceded that the court could still refer to the Wife’s affirmations despite the non-compliance with the rules, as the court’s approach was inquisitorial in nature. He cited a short passage from the judgment of Karminski J in Trotman v Trotman (1964) 108 Sol. Jo. 159:

‘The probate court had always done its best to elicit the facts of the cases it had to try without too rigid an adherence to the rules of evidence practised in other courts.’ (p. 160)”

324.  In my view, the appellant’s affirmations of testamentary scripts are to be treated as having been admitted as evidence of their contents either by implied agreement at the trial or by express concession on appeal.  Insofar as Yuen JA may have held otherwise, I am respectfully unable to agree.

J.2     The weight to be given to the affirmations

325.  The affirmations therefore constitute evidence that the envelope later found to contain the 1990 Will was given to the appellant by Mr Wang and that until the envelope was lodged in court she had kept it in her safe.  The affirmations are therefore important evidence supportive of the genuineness of Mr Wang’s signatures.  But questions arise as to the weight they are to be given.

326.  Mr Thomas submitted that if the affirmations were admissible, the fact that there was no cross-examination on them means that they must be given no or next to no weight.  Mr Vos countered that such a criticism did not avail the respondent since he could have applied to Yam J to order cross-examination but, having failed to apply, could not now belittle the evidence for lack of cross-examination.  Mr Thomas questioned the power of the court to order cross-examination and contended in any event, that any failure to apply for cross-examination could only have been due to a misunderstanding and could not be taken as acceptance of the truth of the affirmations.  An inadvertent omission should not, he argued, be used to turn those affirmations into conclusive points decisive of the case. 

327.  In my view, there is merit on both sides of this debate.  I have held that the affirmations of testamentary scripts must be treated as having been received as evidence of their contents, in particular, as to the provenance of the 1990 Will.  It is also in my view clear that the court could have been asked to order cross-examination of the appellant on those affirmations, so that the fact that they have not been tested in cross-examination should not be held against the appellant so as to deprive them of all weight.  At the same time, the fact is that they have not been tested.  They obviously cannot be accepted as conclusive of the matters deposed to.  Their value as evidence must be assessed, weighing them against any adverse evidence and findings.  It is to such evidence and to the arguments attacking the appellant’s credibility that I now turn.

K.      Attacks on the appellant’s credibility

K.1    Delay in lodging the 1990 Will 

328.  Yam J accepted as a point damaging to the appellant’s credibility, what he took to be unjustified delay in the filing of her first affirmation of testamentary scripts and her lodging of the 1990 Will with the court. 

(a) He found that she:

“...... had shown extreme reluctance to disclose to the plaintiff that she was in possession of the 1990 Documents which she knew were the ‘Wills’ and testamentary directions of Wang.” (§5.115)

(b) Her explanation that she was resisting the opening of the envelope because of her concern for Mr Wang’s safety and out of respect for his wish that it should not be opened during his lifetime (believing that he was still alive), were described by Yam J as “daft and illogical” in a passage copied verbatim from the respondent’s final submissions.

(c) The delay, in his view, justified two highly damaging inferences which clearly imply that the appellant was party to forging Mr Wang’s will.  The inferences which Yam J drew were that:

“(a) the defendant hoped that by delaying and hopefully stifling the plaintiff's action, there would never come the time when the envelope would have to be opened; and

(b) the defendant knew that the contents of the envelope could not be such that could put any person’s mind to rest that it was a valid will and hence even if they were shown to the plaintiff, the plaintiff would surely not have accepted that the ‘Wills’ therein were genuine and valid.” (§5.117)

329.  Yeung JA endorsed Yam J’s views, commenting that the appellant had “only produced ‘the 1990 documents’ reluctantly and many years after Wang’s disappearance and after the safety deposit box was opened” and describing her aforesaid explanations as “highly unconvincing, if not outright ridiculous”. (§591, §595)

330.  I do not accept those views.  In the first place, it is remarkable that the genuineness of the appellant’s concerns for Mr Wang’s safety and her respect for his expressed wishes should be ridiculed at trial and on appeal when, on 14 May 1998, Yam J himself had made a ruling which fully accepted both concerns as well-founded, resulting in his order that the envelope not be opened without leave of the court.  He stated:

“In respect of the defendant’s second summons, I also agree with the defendant that there should be a right to confidentiality and privacy in relation to the testamentary disposition of any person until such time when he is dead or has been presumed dead.  Moreover it is the expressed wishes of the testator that the contents of the envelopes should remain un-open until after his death.  Should such an envelope be opened and the testator was then proved to be alive, there may well be irreconcilable consequences depending on what the testator had written in his testamentary script or disposition.

I also accept the defendant’s submission that there may well be some risks as to the safety of the testator and such fears are not illusionary bearing in mind that this matter is concerned with unpredictable and illogical kidnappers. There may well be some information or effects which will push the kidnapper to terminate all desires to extort  money from the family of the kidnapped.  Again I propose not to go into more detailed analysis here.”

331.  It was not suggested at the trial that anything had changed touching on the genuineness or reasonableness of those concerns during the period in question, namely during late 1997 and early 1998.  I therefore find it impossible to see any basis for those damning comments (copied verbatim by Yam J) in November 2002.  It is not surprising that this volte face, based on a passage copied verbatim from the respondent’s written submissions,is relied on by the appellant as evidencing a lack of independent judicial thought in the preparation of Yam J’s judgment.

332.  Secondly, the suggestion that the appellant had delayed inordinately in filing her first affirmation of testamentary scripts fails properly to take into account the procedural history of the proceedings.  The consistent position taken by the appellant was that her husband was still alive and being detained by kidnappers.  She therefore resisted both the respondent’s application for leave to swear death and his proceedings for probate of the 1968 Will. 

333.  When the respondent decided to discontinue his first swear death application on 2 September 1997, the appellant applied to Yam J to dismiss the first set of probate proceedings as an abuse of process on the ground that it could not proceed without leave to swear to the testator’s death.  This was rejected by Yam J but subsequently held by the Court of Appeal to be correct.  Her resistance to filing the affirmation of testamentary scripts was therefore based on a legal position she was taking which was vindicated on appeal.  I cannot see how her resistance in these circumstances can possibly justify drawing the damning inferences mentioned above.

334.  When rejecting the striking-out application on 6 January 1998, Yam J made an unless order requiring her to file her affirmation of testamentary scripts.  The appellant duly complied and lodged the sealed envelope together with her affirmation in court on 16 January 1998.  This was done while her subsequently successful appeal was pending.  There was no inordinate delay.

K.2    The appellant’s illicit knowledge of the contents of the sealed envelope

335.  The respondent argues that in two separate instances, utterances by the appellant betrayed the fact that she must have known the contents of the as yet unopened envelope notwithstanding the statement in her affirmations that she had never opened it, showing that she was lying.  If this contention is made good, then the weight to be attributed to her affirmations must obviously be drastically affected.

K.2(a)         The appellant’s affirmations

336.  In her affirmation of testamentary scripts filed on 16 January 1998 (and later incorporated in her affirmation dated 18 November 1999) the appellant deposed to Mr Wang having handed her the envelope informing her that it contained a new will that he had made and confidential instructions regarding his estate, telling her not to open it until after his death.  She stated that on the basis of what he had said she believed that its contents referred to highly private and confidential matters relating to the disposition of his estate after his death and that she had not opened the envelope. 

337.  On 12 June 1999, the respondent issued proceedings for appointing receivers over Mr Wang’s assets and simultaneously took out a motion for appointment of interim receivers.  This was at a time when the second application for leave to swear death had not yet been heard. 

338.  In her affirmation dated 15 July 1999, the appellant made two main arguments to oppose such appointment.  The first was that the court might refuse leave to swear death, in which case there would be no question of any probate proceedings and so no basis for receivers, interim or otherwise.  The second was that even if leave to swear death was granted, she believed that the will she was propounding, under which she believed she was the beneficiary – and not the will propounded by the respondent – was Mr Wang’s last will, again removing the basis for the appointment of any receivers.  The important passages in this affirmation (with italics supplied) are as follows:

“Even if the Plaintiff succeeds in obtaining leave to swear to the alleged death of my husband, I shall object to his application for a grant of Probate ...... on the basis that his purported 1968 Will has been cancelled, and additionally that I believe that I am the beneficiary of his later Will ......” (§8)

“...... there is a vigorous dispute as to whether or not the Plaintiff is indeed entitled to any grant to my husband’s assets even if my husband is found to be dead (which I do not accept).  It is my belief that the purported 1968 Will has been cancelled.  Furthermore I believe I am the beneficiary of my husband’s last Will which supersedes the purported 1968 Will relied on by the Plaintiff.” (§51(a))

“...... if receivers are appointed without any security (as the Plaintiff proposes) and it then turns out that the Plaintiff cannot prove the alleged 1968 will as well as his entitlement to my husband’s assets because the sealed envelope given to me by my husband contains a later will appointing me as the sole beneficiary of my husband’s assets, what then?  Who will pay for the damage and loss to my husband’s assets as a result of the appointment of receivers? ......” (§51(e))

“If necessary, and if the Court now desires it, I am willing to give an undertaking to the effect that I will ...... take all steps to preserve my husband’s assets until such time as all the matters referred to above have been determined.  ...... In any event, it is my belief that I am the sole beneficiary to my husband’s assets under the terms of his last Will. ...... But even if I am wrong on this point (which I do not accept), I am advised that I have a substantial claim against my husband’s assets under the Inheritance (Provisions for Family and Dependants) Ordinance Cap 481.” (§51(m)-(o))

339.  The argument made by the respondent is that the appellant cannot have known that she was a beneficiary, much less the sole beneficiary, under the will contained in the as yet unopened envelope unless she had lied about not having opened it and not knowing its precise contents.  As Mr Thomas put it:

“It is precisely the sort of exposed untruths, exposed lie, that calls into question all other arguments presented by the person who has told that untruth.”

340.  However, in my view, that point is not made good.  First, if one assumes (as one must in order to test the argument) that Mr Wang handed the envelope to the appellant for safekeeping until after his death, telling her that it contained his will and instructions concerning his estate, it would be entirely reasonable for her to believe that she was at least the principal, if not the sole, beneficiary thereunder.  There were no children of the marriage and, as discussed below, such a belief was consonant with the closeness and mutual affection that the couple had enjoyed.  That she had such a belief is quite consistent with the envelope having never been opened.

341.  Secondly, the italicised words in the passages quoted above make it clear that the appellant was not asserting that she knew that she was beneficiary or sole beneficiary but asserting her belief that she was, and inviting the court by way of argument to take account of the highly damaging consequences of appointing interim receivers if her belief should prove to be right.  Thus:

(a) In §51(e), the reference to “sole beneficiary” is in the rhetorical question asking what would happen if receivers were appointed without giving security and if it should then turn out that she was the sole beneficiary under the will.  In other words, she was advancing as an argumentative hypothesis her potential status as a sole beneficiary, and not thereby suggesting that she knew that she was such a beneficiary.

(b) In §51(m)-(o), the appellant asserts her belief that she is the sole beneficiary but then expressly accepts that she may be wrong about this.  She goes on to say that in such event, she still has a substantial statutory claim.  This cannot be read as anything other than an argument putting forward what are believed to be reasonable hypotheses.  It does not “expose lies told”.

K.2(b)         The appellant and Lee Kin Sang 

342.  It will be recalled that the envelope was unsealed by Yam J on 6 September 1999.  In the course of that day there were several short adjournments.  Allan Leung’s evidence was that he telephoned the appellant reporting on developments during those adjournments, mentioning the Lee signature on Document A.  His recollection was that the appellant had responded by telling him that Lee was already dead without herself having yet seen the document. 

343.  Mr Thomas pointed to the fact that when asked to decipher the Lee signature in the witness-box, Allan Leung could only say that it looked like “Lee something SG”.  Therefore, it was submitted, Allan Leung could not possibly have supplied the appellant with information that could have led her to conclude who the author of the Lee signature was much less with information allowing her to conclude that such person was dead.  The inference, it was argued, therefore had to be that the appellant had previously seen Document A and thought the Lee signature decipherable as the signature of Lee Kin Sang whom she knew was dead, and that she had given herself away as a liar by prematurely declaring her knowledge. 

344.  Accepting a similar argument, Yam J made a most damaging finding (§§5.104-5.106), stating (of the appellant):

“She could not have known that unless she had already seen Document A or otherwise was a party to the forgery of Document A.”

345.  The abovementioned argument and Yam J’s finding rest on a faulty evidential basis and cannot be accepted.  It is far more likely that Allan Leung’s recollection in the witness-box as to timing was deficient.  The fact of the matter was that Yam J had caused the tape recording of the proceedings on 6 September 1999 to be checked and stated what was found on the tape as follows:

“When I opened the envelope, I said, ‘We have also an English signature, something like “Lee” -- I used the Chinese character Lee – KS, something.” 

346.  Yam J speculated that Allan Leung “might not have picked this up”.  But, since Allan Leung, along with everyone else in court that day, would undoubtedly have been listening intently to what the judge was saying about the contents of the envelope, the overwhelming probability is that he would have picked it up and reported to the appellant that the document bore a signature which, according to the judge, looked like “Lee KS”.  If that is what the appellant was told during one of the adjournments, that would have been sufficient to enable her reasonably to conclude that it had to be Lee Kin Sang, a construction contractor well-known to Mr Wang and herself, who had had business dealings with Chinachem and who had an office just downstairs from Mr Wang.  Her reaching that conclusion and coupling it with her knowledge that Lee Kin Sang had already died would in such circumstances not provide any basis for drawing a sinister inference.

347.  There is an additional problem with Yam J’s finding.  In his witness statement, Allan Leung stated:

“Subsequent to obtaining copies of the contents of the envelopes from the Court and returning to my office on 6 September 1999, I spoke again with Mrs Wang by telephone and made known to her that statements and/or affirmations should be obtained from the two witnesses of the 1990 Will of [Mr Wang] as a matter of urgency setting out the circumstances in which they witnessed the will.”  (Italics supplied)

348.  This suggests that he had not in fact been told by the appellant during an adjournment earlier in the day that the author of the Lee signature was dead.  If he had been told, he would hardly have been suggesting that evening that they should obtain statements from both witnesses. 

349.  The position had changed by 9 September 1999 since in a letter of that date written by Allan Leung to the appellant, reference was made to a need to interview Mr Tse alone.  The inference is therefore that by 9 September, Allan Leung would have been told by the appellant that Lee Kin Sang was dead, but that would not be evidence of any illicit knowledge of the sealed envelope’s contents since it is clear that Joseph Poon had sent copies of the 1990 documents to the appellant after the court rose on 6 September.  Her conclusion that the second witness was Lee Kin Sang and her assertion that he was already dead could therefore at least equally be explicable as having been based on her examining the copy document supplied to her.

K.3    Dr Kung and holiday wills 

350.  As discussed further below, an important issue between the parties concerned the probabilities of Mr Wang making a homemade will.  With a view to promoting such a probability the appellant deposed in her affirmation of testamentary scripts dated 18 November 1999 to having been informed that Mr Wang had made holiday wills which were presumably also homemade.  She stated:

“I should also mention that I have been informed by Andrew Wong to the effect that in or about 1988 or 1989 Mr Wang had told him that he had during those years made a will before going away on holiday.  I have also been informed by Ng Shung Mo that between about 1985 and 1989 Mr Wang made at least one other will which he handed in an envelope to Ng Shung Mo, telling him that if I returned alone to hand the envelope to me and if both Mr Wang and I did not return then to open it himself.  I understand from Ng Shung Mo that when Mr Wang returned safely to Hong Kong from a Chinese New Year holiday in the same year Mr Wang took back that envelope from Ng Shung Mo.  This was plainly prior to the execution by Mr Wang of the 1990 Will ......”

351.  No such holiday wills had been mentioned in the appellant’s original affirmation of testamentary scripts dated 16 January 1998.  Neither Andrew Wong nor Ng Shung Mo were called to give evidence. 

352.  However, a different attempt to establish that holiday wills were used came via the appellant’s brother Dr Kung.  He stated in the witness-box that in the second half of 1997, after having read press accounts which made no mention of a will in favour of his sister, he had telephoned the appellant to remind her that in 1984 and 1985, while he was travelling in a car with the Wang couple, they had on two occasions given him an envelope telling him that it was a will relating to both of them.  His evidence was that the appellant testily told him on the telephone that she knew about those incidents.

353.  However, Dr Kung had made no mention of these holiday wills in his first witness statement of 9 January 2001, raising them only in his second statement dated 28 July 2001.  And as Mr Thomas pointed out, if he had indeed reminded the appellant (who indicated that she did not need reminding) of those holiday wills in the latter part of 1997, it is inexplicable that she did not mention them in her affirmation of testamentary scripts of 16 January 1998.  Neither were they mentioned in her affirmation of 18 November 1999, although there, she mentioned the holiday wills referred to by Andrew Wong and Ng Shung Mo.

354.  Yam J found “that this whole episode of holiday wills in the evidence of Dr Kung is a complete fabrication and should be ignored.” (§5.34)  He had the advantage of seeing Dr Kung in the witness-box and, in my view, he was entitled to reach that conclusion.  The implication is that the appellant was willing to “help along” her case by introducing fabricated evidence to suggest that the homemade will of 12 March 1990 was not the only homemade will prepared by Mr Wang.  This damages the appellant’s credibility and necessarily diminishes the weight to be given to her affirmations of testamentary scripts. 

K.4    Failure to identify the author of the Lee signature 

355.  The majority in the Court of Appeal both considered that the appellant’s failure to identify the author of the Lee signature, or at least to make major efforts to do so, was highly suspicious and damaging to her credibility. 

(a) Yeung JA criticised her for not adducing expert handwriting evidence on whether the signature was that of Lee Kin Sang (§510), the contractor downstairs, since deceased. 

(b) He also criticised her for not searching for the author (and indeed, for the persons who had helped write the text of Documents A, B and C).  He suggested that:

“A simple advertisement could have done the trick—after all, not many people had the opportunity of writing a will and similar documents for a multi-billionaire or witnessing his execution of them.” (§512)

He commented:

“No one who wrote any of ‘the 1990 documents’ or witnessed their execution could have forgotten about the events or the contents of the documents.” (§513)

(c) He concluded that her failure to take such steps in relation to the 1990 documents cast “a grievous doubt on their due execution” (§517).

(d) Yuen JA considered the appellant’s stance as to the identity of the author of the Lee signature to have been inconsistent (§985).  She also stated that if there was doubt as to that person’s identity, the court would expect the appellant to leave no stone unturned in trying to trace that person, but that, instead, the court had been treated to a “startling choice of inactivity” for which there was no satisfactory explanation.

356.  To assess the validity of this attack, the evidence as to the appellant’s attitude and the reasonableness of the steps she in fact took must be examined.  It should be borne in mind that the 1990 Will, being a Chinese will, signature in the presence of attesting witnesses was not an essential requirement.

(a) As previously discussed, at some time after the envelope was unsealed on 6 September 1999, the appellant had expressed the opinion that the signature was that of Lee Kin Sang who was already dead.  I have commented that this was an entirely plausible view given that the signature is reasonably capable of being deciphered by a lay person as the signature of “Lee Kin Sang”, because of Mr Wang’s connections with that gentleman and because of the proximity of his office.  

(b) However, as the appellant subsequently made clear, it was not a signature that she recognized.  This was the basis upon which, in July 2001, the appellant successfully resisted an attempt by the respondent to administer interrogatories concerning the identity of the author of the Lee signature.  On that occasion, Mr Daniel Fung SC informed the court on her behalf that such was indeed her position.

(c) On 4 August 2001, the appellant’s solicitors wrote to the respondent’s solicitors in the following terms:

“In the light of the exchange between Daniel Fung SC and [Yam J] in court on 5th July 2001 concerning the signature in English of the witness on the 1990 will, our client has undertaken further searches and the documents containing the signatures of a Mr Lee that have now been disclosed were located.  We write to give you notice that it is not asserted by our client that she is able to personally lead evidence that the Mr Lee who executed the documents that have now been disclosed is definitely the Mr Lee who was the witness to the 1990 will.  However, it appears that the signatures are similar and accordingly, we consider that disclosure should be given and accordingly, the same has occurred.”

357.  Accordingly, it is not correct to say that she took no steps to ascertain the identity of the author of the Lee signature.  She plainly undertook a search for documents containing known signatures of Lee Kin Sang to compare them with the Lee signature on the 1990 Will.  Having obtained such documents, she again expressed the opinion that the known signatures were similar to the Lee signature on the 1990 Will.  However, whether on the basis of her own expert advice or otherwise, she did not feel able to assert positively that the Lee signature was in fact signed by Lee Kin Sang.  Accordingly, copies of the documents she had obtained were disclosed to the respondent.  Mr Martin Lee SC, appearing for the appellant at the trial stated:

“My learned friend, of course, is welcome to use the information now made available to him on further discovery.  There are experts available to him.  That is entirely a matter for my learned friend.  But I cannot now exclude the possibility that we would never lead any evidence as to what that signature came from.  At the moment, of course, I have no evidence.”

358.  In the first place, I do not think the appellant’s failure to adduce expert evidence can possibly be a matter for criticism.  The documents containing known signatures of Lee Kin Sang were disclosed to the respondent with an invitation to do with them as he wished.  He could obviously himself have adduced expert evidence on the basis of those documents if he thought that it might assist his case.  Presumably, such evidence was not considered helpful.

359.  Secondly, there is no inconsistency between the appellant having (and retaining) an opinion, based on reasonable grounds, that the signature was probably that of Lee Kin Sang, while at the same time stating that she did not personally recognize it and did not feel able to assert positively that it was his signature.  Indeed, on the available evidence, the court would be entitled to find, on the balance of probabilities, that it was the signature of the deceased Lee Kin Sang, a finding which I would be inclined to make.

360.  The import of the criticism regarding failure to make efforts to identify the author of the Lee signature such as by advertising in the press is unclear.  If the suggestion is that some bona fide witness might be alive and might be found by a “simple press advertisement” and so be able to settle once and for all the controversy then, with respect that suggestion lacks realism.  As Yeung JA acknowledged, such a witness would hardly be likely to have forgotten that he or she had witnessed the 1990 Will.  It is overwhelmingly likely that any such person would be aware of this extremely high-profile piece of litigation and could be expected to have come forward without the need for a press advertisement. 

361.  If, on the other hand, the criticism implies that the appellant failed to investigate because she knew, as part of the conspiracy, that there was no true author of the Lee signature, such an inference is not justified.  The appellant’s conduct is quite consistent with her reasonably believing it probable (on the grounds already mentioned) that the witness was the deceased Lee Kin Sang.  Such a belief provides a reasonable explanation for her not doing more to find the person signing the Lee signature.  The fact that Lee Kin Sang had died on 21 June 1996, before commencement of the proceedings, also provides a reasonable explanation for the fact that no one had come forward claiming to be the author of the Lee signature. 

362.  Accordingly, while, as Mr Vos accepted, one is able to suggest various further measures that the appellant might have taken to investigate the Lee signature, I am unable to conclude (especially  bearing in mind the Re H and Lee Ming Tee principles) that her failure to take such steps damages her credibility or lends any weight to the forgery conspiracy theory.

K.5    The absence of oral testimony from the appellant and the overall status of her evidence in the case

363.  It is no surprise that a great deal of emphasis was placed by the respondent on the fact that the appellant did not give evidence in the witness-box.  Yeung JA described both the statements of Mr Tse and the appellant’s affirmations of testamentary scripts as “untested by cross-examination” and as leaving “a lot of unanswered questions”, concluding that little weight could be attached to them.  Yuen JA apparently did not consider the appellant’s affirmations to be in evidence.  My conclusion as to the status of Mr Tse’s evidence is set out in Section I.4 above and repetition is not required.  However, in considering the appellant’s position, it is important to distinguish two distinct areas in which her evidence is or may be important. 

364.  In the first place, she had important positive evidence to give regarding the provenance of the will: evidence that Mr Wang had handed it to her in a sealed envelope in 1990 and that she had kept that envelope safely and unopened until it was lodged in court in 1998.  That evidence was given in her affirmations of testamentary scripts, and, as discussed in Section J above, was received in evidence at the trial, subject to questions of weight (which are presently under discussion).  Accordingly, the fact that the appellant did not enter the witness-box does not mean, as it otherwise might, that there is no evidence about the provenance of the will.  And while the weight of such evidence must be assessed in the light of challenges to the appellant’s credibility, the fact that she was not cross-examined on those affirmations cannot in itself mean that no or only negligible weight should be attributed to them, given that it was plainly open to the respondent to call for cross-examination which, probably by mistake, he had failed to do.

365.  Having considered in this Section various attacks on the appellant’s credibility, it is my view that her credibility did sustain damage by reason of the court’s findings concerning the holiday wills.  The weight to be given to the affirmations of testamentary scripts was therefore plainly diminished.  Such weight was obviously not enhanced by the fact that there was no cross-examination, although that point is attenuated by the respondent’s mistaken omission to require any cross-examination.

366.  If the case rested solely on the appellant’s untested affirmations of testamentary scripts, it may very well be that the court would refuse probate of the will propounded by the appellant.  However, that is not the position in this case.  Those affirmations occupy a position of secondary importance in relation to Mr Tse’s statutory declaration and affirmation, which are direct evidence of Mr Wang’s execution of the 1990 Will.  I have held that the Tse statements weathered the attacks made on them and emerged largely unscathed.  The appellant’s affirmations are essentially corroborative of those statements and they take their significance from the cogency of those statements.

367.  The second area in which the evidence of the appellant is or may be important involves refuting the respondent’s case on forgery and on the conspiracy involving the subtle form of forgery.  Adverse inferences can only be drawn from a failure to testify in such a context where there is a need to meet an established prima facie case.  In M’Queen v The Great Western Railway Company (1875) LR 10 569, Cockburn CJ stated:

“If a primâ facie case is made out, capable of being displaced, and if the party against whom it is established might by calling particular witnesses and producing particular evidence displace that primâ facie case, and he omits to adduce that evidence, then the inference fairly arises, as a matter of inference for the jury and not as a matter of legal presumption, that the absence of that evidence is to be accounted for by the fact that even if it were adduced it would not disprove the primâ facie case. But that always presupposes that a primâ facie case has been established; and unless we can see our way clearly to the conclusion that a primâ facie case has been established, the omission to call witnesses who might have been called on the part of the defendants amounts to nothing.” (at 574)

368.  As discussed above, before a prima facie case of forgery has to be faced by the appellant, the respondent must have adduced sufficient evidence probative of forgery in accordance with the Re H standard and the Lee Ming Tee principle.  In this context, Gilliard J’s statement in the Supreme Court of Victoria in O’Donnell v Reichard [1975] VR 916 is apposite.  His Honour stated (at 920):

“...... any party upon whom the burden of proof on any issue is imposed must always adduce a prima facie case on such issue to go to the jury, and the failure of the other party to the litigation to call witnesses who may be expected to elucidate the matter cannot fill any gaps in the proof required (see per Dixon CJ in Hampton Court Ltd v Crooks (1957) 97 CLR 367 at p 371; and Tyne v Rutherford (1963) 36 ALJR 333).”

369.  It follows that no question of drawing adverse inferences arises unless the respondent adduces evidence of a sufficient cogency to raise a prima facie case of forgery and of the implied conspiracy.  Only then would the appellant need to refute that case and only then could the court be invited to draw adverse inferences from the absence of oral testimony from the appellant.  But if the respondent does not cross the relevant evidential threshold, her omission to testify, as Cockburn CJ noted, “amounts to nothing”.  And her omission to testify does not, as Gilliard J pointed out, help the respondent to fill any gaps in his forgery and conspiracy case.

370.  The importance attached by Yam J and the majority in the Court of Appeal to the omission to testify is plainly linked to the fundamentally flawed approach of requiring the appellant to “dispel suspicious circumstances”.  On that approach, many unanswered questions, characterised as undispelled suspicious circumstances, are bound to remain where the appellant has not gone into the witness-box to dispel them.  However, whether or not the appellant testified in the witness-box, many questions identified as “suspicious” by the respondent would necessarily have remained unanswered.  How could she have been expected to explain, for example, why there were two different writers for Documents A, B and C?  Or the idiosyncratic language in Documents B and C?  The absence of answers to questions such as these merely reflects the impossible demands of the erroneous burden.  It provides no basis for criticising the appellant.

L.      The propensity of Mr Wang to make a will like the 1990 Will 

371.  Acceptance of the evidence of Mr Tse as directly probative of execution of the 1990 Will by Mr Wang and acceptance of the appellant’s affirmations of testamentary scripts as corroborative evidence of provenance effectively disposes of the case.  Evidence called with a view to suggesting that the 1990 Will was not the sort of will that Mr Wang could be expected to make is inherently of a less compelling order.  As Viscount Haldane, LC said in Chetty v Chetty (1916) AIR 113 (PC):

“When a will has once been made and is apparently in perfect form, and the evidence of the attesting witness is to be trusted, few things can be more dangerous than to attempt to recreate the kind of will that the man ought, in the opinion of the Court, to have made.” (at 115)

372.  The handwriting evidence finds itself at a similar disadvantage by comparison with direct evidence of execution.  However, out of deference to the views taken in the courts below, I will address some of the principal arguments accepted below. 

M.     Choice of the appellant as beneficiary 

373.  Yam J concluded that the evidence disclosed no reason for Mr Wang to change his intention to benefit the respondent exclusively, as expressed in the 1968 Will.  This he took as his first “suspicious circumstance” to be held against the 1990 Will.

374.  That finding is unfounded and runs contrary to the evidence.  Yam J (at §5.12) supports his conclusion by suggesting that “there is no doubt that Teddy made his fortune because of the business he took over from his father”.  But the evidence is quite to the contrary.  Mr Wang made his fortune not out of taking over his father’s chemicals business, but after he and the appellant struck out on their own as property developers during the early 1960s, building up their empire together in the thirty-odd years prior to Mr Wang’s disappearance. 

375.  Yam J also suggests that his conclusion is supportable by evidence that “...... Teddy always had a deep gratitude and respect for his father for what he had provided to him especially at his tender age”(§5.14), pointing to the respondent’s having kept secret the fact of Mr Wang’s expulsion from school in 1952.  It is telling that Yam J felt it necessary to base his inference regarding Mr Wang’s testamentary intentions in 1990 on an incident occurring 38 years earlier during Mr Wang’s schooldays.  So far as the “deep gratitude and respect” for his father was concerned, it has to be said that this was hardly reflected in the allowance of $19,000 per month given to his father, of which $8,000 had to go to his mother.  Given the extreme wealth of Mr Wang, such a sum, from anyone less notoriously frugal, would have been described as derisory.

376.  Yam J also relied heavily on the period of serious matrimonial problems encountered by the Wang couple in about 1968 (§4.20).  Indeed, he went so far as to make a finding that the appellant had committed adultery, criticising her because there “was never any statement of denial made on oath by Nina of her extramarital relationship with someone at that time”(§3.35).  That finding was wholly unwarranted.  Whether or not there was actual adultery was never in issue and it was no part of the judge’s function to make such a finding.  There was no basis for criticising the appellant for failing to deny this non-issue on oath.  It is impossible to see what relevance such a denial would have had in any event.

377.  Notwithstanding the undoubted matrimonial problems which had led to the making of the 1968 Will, the evidence overwhelmingly established that the couple had become fully reconciled by the mid-1970s and that they had thereafter been inseparable as business partners working to build up their empire (which, being childless, they liked to refer to as “their baby”).  It also established that they had regularly been seen in public as a close and affectionate couple.  Yam J himself acknowledged that the evidence of a number of witnesses, including Donald Cheung (one of the respondent’s own witnesses), was that they were a “‘loving couple’ cum business partners” (§4.3). 

378.  Finally, Yam J sought to support his conclusion that the appellant was an unlikely choice of beneficiary because he had already “made adequate provision” for her by inter vivos gifts (§4.21), concluding:

“In the end, it is fair to conclude that the improved good relationship and generous inter vivos provisions would only indicate that Teddy would like to benefit Nina by inter vivos provisions only and not by testamentary provisions.” (§4.37)

379.  That is an unsupportable conclusion.  There was no evidence of any inter vivos gifts by Mr Wang to the appellant.  The evidence showed the structuring and restructuring of the corporate empire and the allocation of shareholdings therein, not gifts from one to the other.  There is in any event not a shred of evidence that Mr Wang had chosen to provide for the appellant by inter vivos and not by testamentary means.

380.  To conclude that there was no reason for Mr Wang to change his testamentary intentions was wholly untenable.  The first and immediately obvious reason for a change is that the motivation for making the 1968 Will, namely, the serious matrimonial problems which had occurred 22 years previously, had long been left behind.  Moreover, in 1968, Mr Wang was aged 33 and his father was 56, still working and commercially active.  But in 1990, Mr Wang was himself aged 56 and his father aged 79.  By then he, in partnership with the appellant, had built up a phenomenally successful property empire in the course of a marriage that had lasted some 35 years. It was far more likely that he would leave his portion of that empire to the appellant who had built it with him (which they referred to as “their baby”) and who would be able to continue operating it, than leave it to an octogenarian who had already been in retirement for 13 years.  Evidence of discussions which the Wang couple had with financial and legal advisers between 1987 and 1990 with a view to setting up a tax shelter plan indicates that all such planning envisaged Mr Wang and the appellant as the ultimate beneficiaries, with no mention of the respondent in these arrangements.  All of these are matters which make it entirely unsurprising that the appellant should be named as sole beneficiary.

381.  Yam J’s views did not receive support in the Court of Appeal.  However, Yeung JA, adopting a less extreme position, suggested that there was “no reason for Wang to completely sever the Father from his estate” (§559).  In the light of the considerations discussed in the preceding paragraph, I am unable to agree.  Moreover, Document B contains a precatory direction that the appellant should “take care of” Mr Wang’s parents which indicates an intention that the modest maintenance payments which had been made ought to be continued.

382.  Yuen JA dismissed the suggestion that naming the beneficiary was a “suspicious circumstance”.  She held that both the appellant and the respondent were natural legatees (§908).  As indicated above, my view is that the evidence goes further and favours a finding that the appellant was far more likely to have been the chosen beneficiary. 

N.      Other “suspicious circumstances”

383.  There was therefore nothing suspicious in Mr Wang’s choice of the appellant as his sole beneficiary.  Once this is accepted, the other allegedly “suspicious circumstances” fall into perspective.  Those suspicions do not involve challenging the idea that the estate should go to the appellant, but only arguments questioning why it took so long for Mr Wang to change his will and why when he did, he chose to do so in such an informal manner.  These are significantly less compelling concerns.

N.1    Delay in revoking the 1968 Will 

384.  It is indeed surprising that Mr Wang waited until 1990 before he got around to changing his 1968 Will, given the couple’s reconciliation and the other features of his relationship with the appellant discussed above.  However, it is an everyday observation that people often procrastinate when dealing with their own personal affairs, even though they may handle other people’s affairs with great professional efficiency.  This is perhaps understandable in relation to wills since many may prefer not to contemplate their own deaths and may shrink from making testamentary dispositions with their own deaths in mind.

385.  The evidence, in my view, makes it probable that the riding accident on 10 March 1990 jolted Mr Wang into action.  There was much debate at the trial as to whether the accident was or was not serious.  Plainly, it was not particularly serious, but neither was it trivial.  Mr Wang suffered a concussion, a loss of consciousness, a loss of memory and various cuts and bruises that put him in hospital for a CT scan and a stay of two days before he discharged himself against medical advice.  There was evidence from his godson that he was frightened by the experience of losing his memory.  But the degree of seriousness of his injuries is not to the point.  It is sufficient to find it probable (as I do) that Mr Wang realised that the accident could have been much worse, prompting him to make the will upon leaving hospital.

N.2    Why a homemade will? 

386.  Assuming that the making of a homemade will was an immediate stop-gap measure, having been stung into action, one may fairly ask why Mr Wang did not have it replaced by a formal will drawn up by solicitors along the lines of the 1960 and 1968 Wills.  There is no explanation evident from the evidence.  However, it must be kept in mind that Mr Wang was kidnapped less than a month later on 10 April 1990.  He obviously did not know that was going to happen and a simple explanation may well be that, having put a stop-gap measure in place, he did not consider its formalization a matter of urgency and was unfortunately kidnapped before having got around to it.  In any event, the fact that the 1990 Will was and remained homemade does not justify an inference of forgery or conspiracy, nor does it displace or weaken Mr Tse’s direct evidence of execution.

N.3    Why such unconventional documents and idiosyncratic language?

387.  It is also fair to say that the 1990 documents exhibit unusual and eccentric features and idiosyncratic language which make the documents somewhat bizarre and not readily explicable.  It is odd that four documents were included in the sealed envelope when only Document A had a testamentary effect.  It is odd that Mr Tse was evidently asked to sign as witness on all the documents, including Document D which was quite obviously not testamentary.  The fact that Documents B and C refer to Mr Wang being disappointed by his father and mother is puzzling.  Even more so is the reference in Document C to everyone in the appellant’s family being “disgusting”, especially bearing in mind Dr Kung’s treatment of Mr Wang after his accident.  Other notable features include the fact that the documents were untidily written on thin paper and comprised documents with text written by more than one hand.

388.  Mr Thomas referred to Fuller v Strum at §41, as authority for the proposition that the court can and should ask itself whether the testator was likely to take such an eccentric course.  I agree that the court may properly adopt such an approach.  However, it must be remembered that Fuller v Strum is a case where the court was inquiring into the mind of the testator, asking whether he knew and approved of the contents of the will in circumstances where the person who wrote out the will was a named beneficiary – a classic case of suspicion within the second rule in Barry v Butlin.  In that context, a discussion of the propensity of the testator to use certain expressions or to produce a document in a certain form is of more direct relevance.  In our case, we are concerned with the question of whether the signature is that of the testator where the choice of beneficiary is unsurprising.  Whether the will is expressed in terms, or made in a form, which Mr Wang was or was not likely to have adopted is very much a subsidiary question which is relatively remote from the ultimate issue for the court. 

389.  In any case, the existence of the eccentric features in the 1990 documents plainly cuts both ways in the argument.  Mr Vos, citing the example of Wood v Hitichings (1841) 2 Moo PC 355 at 444-5, submitted that the issue being one of forgery versus genuineness, the court is entitled to ask what the probabilities are that a forger would have chosen to create documents with such bizarre features.  I again agree that this is an approach which is open to the court.  Mr Vos submits with much force that it is inherently improbable that a forger would choose to forge four documents when only one would do, magnifying quite unnecessarily the chance that forgery might be detected.  And, as he submitted, it is equally improbable that a forger would have chosen to give unnecessary offence to both the Wang parents and the Kung family as this could only encourage them to challenge the will as not reflecting the testator’s true views.  The forger is also likely to have avoided any questions that might have been raised regarding untidiness or use of scrappy stationery.

390.  This approach was not accepted by Yuen JA who stated:

“...... it is not for the court to try to speculate why a forger would or would not have done something, as the court has no evidence of the relevant circumstances at the time of commission or omission. (§922)

I am unable to agree with her Ladyship’s view.  It is true that the court has no direct evidence of matters taken into account by the forger; however, the only assumption which the court needs to make for these purposes is the inescapable and wholly justified assumption that a forger wishes his handiwork to succeed, that is, to fool the world into thinking that his signature is in fact that of the testator.  It inevitably follows that the probabilities would weigh against the forger taking unnecessary risks of being found out, such as by forging signatures on four documents when signatures on one document would do, and so forth.  Such reasoning does not involve speculation, but merely an examination of the inherent probabilities.

391.  My conclusion on this point, as noted above, is that the evidence regarding the form and content of the 1990 documents cuts both ways.  It raises unanswered questions which can be argued to favour in turn the appellant or the respondent.  The outcome of the argument on this point therefore does not have a significant impact on where the balance of probabilities lies.

O.      The handwriting evidence 

392.  I have had the advantage of reading in draft the judgments of Mr Justice Chan PJ and Mr Justice Litton NPJ and gratefully adopt their analysis of the handwriting evidence. 

O.1    The nature of the expert handwriting evidence generally

393.  As previously indicated, handwriting evidence is necessarily of a lower order of cogency than direct evidence of execution.  This is illustrated by Fuller v Strum [2002] 1 WLR 1097, where a single joint expert had expressed the uncontradicted opinion that “there was ‘very strong positive evidence’ that the signature on the will was a forgery”.  The trial judge, however, declined to accept that evidence because it necessarily implied fraud on the part of several persons which was a conclusion that he rejected (see §23).  On appeal, his decision was partially reversed on other grounds, but his rejection of the expert’s opinion and of forgery was not questioned on appeal. 

394.  Secondly, it is important, when evaluating expert evidence on handwriting, to recognize that such evidence has both a harder, more “scientific”, side and a softer, less precise, facet involving professional judgment of a relatively subjective nature.  The point is put thus in Phipson on Evidence (Sweet & Maxwell, 15^th Ed, §37-10):

“Expert witnesses have the advantage of a particular skill or training.  This not only enables them to form opinions and to draw inferences from observed facts, but also to identify facts which may be obscure or invisible to a lay witness.  The latter might simply be described as ‘scientific evidence’, the former as ‘expert evidence of opinion’.  A microbiologist who looks through a microscope and identifies a microbe is perceiving a fact no less than the bank-clerk who sees an armed robbery committed.  The only difference is that the former can use a particular instrument and can ascribe objective significance to the data he perceives.  The question of subjective assessment and interpretation which is the essence of opinion evidence hardly enters into the matter at all.  An example of the dichotomy can be seen in the case of a conflict between experts on handwriting as to the authenticity of a document.  By virtue of their training such experts would be able to distinguish parts of letters or techniques of word-formation which a layman would be unable to observe: this is the scientific part of their work.  The question of which features are significant and the inferences to be drawn from them are questions of judgment, assessment, opinion.”

It is worth bearing this distinction in mind when considering some of the approaches adopted below. 

O.2    Features of the handwriting evidence in the present case

395.  In the present case, the evidence of the respondent’s three experts as to forgery of the Wang signatures must be viewed with some reserve.  This flows from the rejection (discussed in Section G.6 above) by two members of the Court of Appeal of their evidence (and Yam J’s finding) that the Tse signatures were forged – evidence and findings which Mr Thomas was unable to support before this Court.  Since those experts had expressed themselves in terms of significantly greater certainty in relation to the Tse signatures, the weight to be given to their evidence on forgery in relation to the Wang signatures must be questioned.  I hasten to add that in saying this, I do not for a moment doubt that all four experts testifying on handwriting – Mr Cheng, Mr Lesnevich, Mr Tsui and Professor Jia – did their best to render their professional assistance in good faith to the court.  It is in the nature of such evidence, particularly in relation to matters of subjective expert judgment, that substantial room exists for disagreement.  I share the view of my colleagues in this Court that the attacks on the honesty and integrity of Professor Jia were unfounded and should not have been made.  I return to that topic below.

396.  The second feature of the handwriting evidence in the present case concerns the limited range of known samples.  It is of course essential for an expert examining questioned signatures to be able to compare them to what are known signatures of the purported author.  It is also obvious that the wider the sample of known signatures, the more reference points the examiner will have against which to consider the genuineness or falsity of the questioned signatures and the more secure his opinion will be.

397.  In the present case, 18 samples of known signatures were produced.  The earliest were two samples from 1958 (“the 1958 samples”), so that the period to cover up to the date of the 1990 documents, was some 32 years.  However, the 18 signatures only went up to 1985, no later samples having been available to the experts.  Moreover, those 18 samples were not evenly distributed.  They came from only seven different years (1958, 1967, 1975, 1980, 1983, 1984 and 1985).  Thus many years were unrepresented and several signatures came from the same year.  There were two samples from 1958, two from 1983 and four from 1984.  Furthermore, it was common ground that the questioned signatures had been written using a broad-nibbed calligraphic pen, a writing instrument that would undoubtedly have some effect on the manner and speed of writing and the resultant signature.  However, no samples written with the same type of pen were available.  The 1958 samples and a sample from 1980 were written using a fountain pen.  The others were written with a ball pen or with aqueous ink which could have come from a roller ball or a fountain pen.  In one case, the sample was written with a felt-tip pen.

398.  It appears to be the case that Mr Wang had come to use his English signature far more frequently than his Chinese signature.  Accordingly, the experts were necessarily hampered to some degree by the limited range of available samples.  And since not all of those samples were immediately available, they had to modify some of their opinions as further samples were produced. 

399.  Thirdly, the appellant’s handwriting evidence initially submitted was contained in a joint report of three experts comprising Professor Jia, Professor Xu Li-gen and Professor Zhan Chu-cai.  That report was, with justification, criticised for superficiality and for reaching some dubious conclusions based on unsatisfactory grounds.  However, only Professor Jia was called to give evidence on behalf of the appellant and his examination-in-chief, quite extraordinarily, lasted for 17 days as he in effect restructured and rehabilitated the expert evidence relied on by the appellant.  He was then cross-examined and re-examined over a further 19 days.  Such a turn of events was regrettable and suggests a deficiency in preparation of the case.  However, it is clear that Professor Jia’s evidence in the box was far more comprehensive and cogent than the contents of the joint report.  He was subjected to cross-examination in minute detail.  Accordingly, while there was understandable criticism of the joint report in the judgments of Yam J and the majority in the Court of Appeal, and while Professor Jia opened himself to some adverse comment by having subscribed to the unsatisfactory report, at the end of the day, the expert evidence adduced for the appellant which must be assessed is primarily that given by Professor Jia in the course of his lengthy testimony and not the evidence in the joint report.

O.3    Yam J’s approach to handwriting 

400.  Yam J’s approach to the handwriting evidence came under severe attack.  He was criticised for abdicating his judicial responsibilities by reproducing virtually verbatim almost the entirety of the first set of written submissions on handwriting evidence lodged by the respondent and ignoring answering submissions filed by the appellant.  It was submitted, with considerable force, that the Court would be justified in treating his findings as not judicial findings at all.

401.  But on the footing that one does treat the judge as having made findings, albeit adopted from the submissions, his approach was criticised on the fundamental ground that he had focused almost exclusively on those parts of the evidence which arguably tended to prove forgery and ignored or rejected on invalid grounds other parts of the evidence which were arguably probative of genuineness. 

402.  In particular, Yam J was criticised for rejecting or ignoring clear evidence that the questioned signatures were written in a free-flowing, natural and rhythmic manner with uninhibited pen pressure on thin paper, all of which were features probative of genuineness.  He was said to have done so on a demonstrably faulty basis devoid of expert evidential support, having purported to detect nine instances of “retouching” which were not detected by the experts. 

403.  Secondly, Yam J was criticised for failing to understand or to give proper weight to the probative value of matching natural variations as explained by Professor Jia.  The existence of natural variations in the way a person signs his signature is acknowledged in the literature.  Accordingly, where more than one questioned signature exists, the examiner will wish to consider any variations detectable in the way in which those questioned signatures were written, and then examine the known signatures to see whether variations existing among those signatures match the variations seen in the questioned signatures.  Plainly, a matter of subjective judgment enters into this exercise.  However, where the matching variations detected (usually by greatly enlarging images of the signatures) are subtle and inconspicuous and where they are personal to the author and not variations commonly appearing in the population at large, they may be strongly probative of genuineness because they are exceedingly difficult to forge, whether by simulation or by tracing.  Moreover, where the examiner is able to find a series of matching variations which exist in both the questioned signatures and the known signatures, the probability of genuineness is augmented since the difficulty of forgery is proportionally multiplied.  Yam J is attacked for failing to comprehend this evidence or, at any rate, failing to give any weight to the multiple matching variations identified by Professor Jia.

404.  Thirdly, Yam J’s acceptance of the evidence providing the basis for his finding of forgery is criticised.  Yam J confidently found that the Wang signatures were forgeries by simulation.  Fundamental to that finding, was his conclusion that the questioned signatures were modelled on signatures similar to the 1958 samples and that such signatures were written in a style that was no longer in current use by Mr Wang by the time he is said to have signed the 1990 Will.  There were in consequence, he found, significant differences between the more current signatures and the questioned signatures. 

405.  The appellant challenges this finding on the basis, first, that there was no acceptable evidence that the appellant no longer signed in a style similar to that adopted in 1958.  Only Mr Lesnevich suggested that Mr Wang had lost the ability to write in such manner.  All the other experts disagreed.  Given the limited number of samples available, it was argued that a finding of forgery said to involve copying signatures similar to the 1958 samples was unjustified.  Secondly, it was argued that such a finding was reached without taking into account the effect of writing with a calligraphic pen.  It was said that the evidence showed that use of such an instrument would result in slower writing and a greater nuancing of particular strokes when compared, for instance, with signatures written using a ball-pen or roller-ball.  In short, the criticism is that the judge was not comparing like with like when concluding that there were significant dissimilarities proving forgery.

O.4    One example examined

406.  All these criticisms are of substantial weight.  In the light of the analysis put forward by Mr Justice Chan and Mr Justice Litton, I will confine myself to one example which illustrates how, in my view, Yam J failed to avoid a number of serious pitfalls.  This example involves the character “Wang” (王) and Yam J’s finding that in the questioned signature on the 1990 Will, the forger retouched the downward vertical stroke in that character at the point where it is intersected by the second horizontal stroke.  Professor Jia did not accept this and, copying verbatim the written submission made by counsel, Mr Edward Chan SC, the professor was castigated for having given a dishonest answer in this context.

407.  Where a questioned signature shows retouching, that is, an attempt by the author to repair or improve the signature by over-writing or adding a stroke, this may be a strong indication of forgery.  It certainly deprives the signature of the naturalness that is a hallmark of genuineness.  However, none of the respondent’s own experts suggested that there was any sign of retouching in any of the questioned Wang signatures.  On the contrary, Mr Lesnevich and Mr Cheng both gave evidence saying that they had not detected any retouching. 

408.  When Professor Jia came to be cross-examined, notwithstanding the absence of any evidential foundation from the respondent’s own experts, Mr Chan put it to him that there were instances of retouching in the questioned signatures.  Mr Lee SC objected and the exchange proceeded as follows:

“MR CHAN:   What I am suggesting is that there was a re-touching over that part. 

HIS LORDSHIP:   I think you have to put that to the Professor. 

MR LEE:   My Lord, this is totally new.  What is the point of having experts’ reports when none of the three experts called by my learned friend said this at all? 

HIS LORDSHIP:   Yes, Mr Chan? 

MR CHAN:   Well, my Lord, we agree that, of course, Mr Tsui, Mr Cheng and Mr Lesnevich did not mention this, and in fact they did not actually discover that at the time when they -- bearing in mind that the only time they did the microscopic examination was back in 2000, and it was really after they had given evidence, when the whole thing was being shown on the microscope ......

HIS LORDSHIP:   So you are just suggesting it to the Professor to see whether he agrees, and that is the end of the matter? 

MR CHAN:   That is the end of the matter.”  (Transcript 96.67.20-68.13)

409.  The suggestion of retouching in relation to the character “Wang” (王) was based on the following reasoning which had already been explored in cross-examination with Professor Jia:

(a) In the questioned Wang signature on Document A, the character is written approximately as follows (the image here not being an accurate reproduction):

(b) In his known signatures, to write this character, Mr Wang would (i) first write the topmost horizontal stroke, (ii) then write the vertical downwards stroke and then, (iii) form the second horizontal stroke by a continuous upward looping movement so as to intersect the vertical stroke crossing from left to right, before (iv) looping downwards again to complete the lower part of the character.

(c) However, since the vertical line in the (王) character in the 1990 Will was seen to be darker than the intersecting horizontal line, it was suggested to Professor Jia that that vertical line must have been written or retouched, after the horizontal line had already been written. 

410.  Professor Jia did not agree.  He explained that in writing the vertical stroke (ii) above, a quantity of ink had been deposited on the paper which made a heavy and thick line, but that when the upward looping stroke (iii) above was written, this was done with a quicker movement and lighter pen pressure so that less ink was deposited on the paper, with the result that the horizontal line appears lighter.  It followed that the fact that the vertical stroke (ii) looked darker did not mean that it had been written on top of and subsequently to the horizontal stroke (iii) or that there had been any retouching.

411.  Notwithstanding his explanation, and with a view to getting him to accept the theory of retouching, Professor Jia was invited to examine a greatly enlarged image of the intersection of the two lines in question through a microscope, it being put to him that it was possible to see, by virtue of a difference in colour tone, the actual retouching stroke.  The image appearing below is an approximation of the computer-captured image viewed by Professor Jia :

412.  The transcript records the following exchanges:

“A.  I have made a direct examination of this photograph supplied by you under the microscope and I cannot see any difference. 

Q.  Of course, under the microscope, what you are looking at is black colour, and any difference in tone in black obviously is very, very difficult to observe?

A.  Yes, it is quite difficult. 

Q.  For some reason, of course, when you are making colour photographs, sometimes you may not be able to make it in the exact tone, and the difference in tone would appear easier to be observed if the tone is lighter; do you agree?

A.  I cannot see much.  I cannot see any obvious difference here on this photograph. 

Q.  I am suggesting to you that there might be -- in fact, there was a re-touching starting right from here (Indicating), down, up to here.  I am indicating to the approximate position. 

A.  No, I cannot see any re-touching.” (Transcript 96.69.12-70.5)

413.  It is noteworthy that when Professor Jia, looking through the microscope, said he could not see a difference in colour tone, the cross-examiner agreed that it was “very, very difficult” to observe such a tone.  And when the professor disagreed with the suggestion that there was retouching, one would have thought that, in line with the earlier exchange with Yam J, that would be the end of the matter.

414.  However, it was not the end of the matter.  The (copied) passage in the judgment of Yam J dealing with this matter (§10.59) states as follows:

“The ink would indicate that the downward movement of the vertical stroke would appear to be on top of the upward stroke. This feature is best shown in F7A-1.  Due to the special lighting effect, this feature is not as clear although it is still visible in the computer capture in A.01.w01 but is not shown in A.01.w04.  This phenomenon would indicate that after the original downward and upward strokes were written (as one continuous stroke) there was some retouching of the downward stroke over the intersection point so that one can now see that the downward stroke covered the upward stroke. The purported explanation from Prof. Jia was that this phenomenon was caused by the fact that the ink in the downward stroke was rather heavy and thick and when the pen went on to write the horizontal stroke, the movement was relatively faster and the pen pressure was lighter, so the ink line for the upward (horizontal) stroke was light in colour and was faint.  In fact when the difference in the colour tone between the downward stroke and the middle horizontal stroke on either side of the downward stroke was pointed out to Prof. Jia, he claimed that he could not see the difference in the colour tone!  This is most surprising and it cannot be an honest answer as such difference in colour tone is obvious from F7A-1 (a photo taken before the cracking occurred or got more seriously in the original Document A).”

415.  That finding and the submission which had led to that finding, were grossly unfair to Professor Jia.  The expert evidence provided no support at all for the existence of the suggested retouching.  A denial by Professor Jia should properly have been the end of the matter, as Yam J and Mr Chan SC themselves had earlier recognized.  In any case, Professor Jia had not denied that in the photograph in question (F7A-1) there was a difference in colour tone between the vertical and horizontal strokes.  On the contrary, as noted above, he had provided an explanation in terms of ink flows and writing movements to explain that colour difference and why it should not be interpreted as showing that the vertical stroke was written in sequence after the horizontal stroke.  When Professor Jia said that he could not see a different colour tone, he was dealing with a different question, namely, whether, looking through the microscope at the hugely enlarged image of the intersection segment of the character (A.01.w01), it was possible to detect the actual stroke used to retouch the character by virtue of a difference in colour tone shown in the enlarged image.  He said that he could not see such a difference and Mr Chan SC agreed that it was “very, very difficult” to observe a difference.  It was never put to Professor Jia that his answer was dishonest.  It is therefore inexplicable how the suggestion of dishonesty came to be made in submissions and how it came to be adopted by Yam J.  In the circumstances, it was a wholly inappropriate allegation which should never have been made.

416.  There is another instructive aspect to this example.  It shows the dangers of a judge purporting to make expert judgments in areas where expertise is needed.  In drawing conclusions without expert assistance about stroke sequence based on his non-expert observation of a darker line intersecting with a lighter line, Yam J made a well-known error which the expert literature cautions against.

(a) Albert S Osborn, “Questioned Documents” (2^nd Ed, Nelson-Hall Co) explicitly warns :

“Special caution is necessary in determining the sequence of lines of different width or different intensity of colour.  A heavy or dark line always seems to be over a light or indistinct line when they cross, and a pale line, or light colored line, always seems to run under a black line at a crossing.  Under these conditions it may be impossible to determine which was last made and the inexperienced and uninformed may easily be led into error.” (Italics in the original, at 510)

(b) Wilson R Harrison, “Suspect Documents” (1958, Sweet & Maxwell) puts it thus:

“The fact that truth is rarely what appears ‘obvious at a glance’ can be demonstrated by considering the well-known optical illusion – that, of two intersecting lines, the darker invariably appears to have been written over the lighter, irrespective of the true state of affairs.” (at 226)

417.  Despite Waung J pointing out this error on appeal (at §1232), it was not corrected by the majority.  Yuen JA did not mention any of the retouchings purportedly detected by Yam J.  Yeung JA was wrongly dismissive of the point, stating:

“Mr. Chang suggested that the indicia of retouching and retracing identified by the judge could be the result of ‘optical illusion’, but the expert Professor Jia did not offer such explanation. The judge could not be criticized for not considering the suggested ‘optical illusion’.

I fail to see any merit in the suggestion that the judge ‘played expert’ as a ground to vitiate any of his findings.”  (§§684-685)

418.  In the first place, Professor Jia did point to a possible optical illusion, explaining that the darker line may be due to the volume of ink in the stroke rather than a different stroke sequence.  Secondly, the judge is certainly to be criticized where his error is due to his striking out on his own into expert territory, contradicting Professor Jia’s evidence without any evidential support for the contrary position and, in consequence, making an error well known to the experts.  His finding of retouching in this example is plainly vitiated by his error.

O.5    Abandonment of the Re H and Lee Ming Tee principles 

419.  In Section H.1 above, I noted that Yam J stated that Re H applies in relation to the handwriting evidence.  However, Yam J’s speculative, “do-it-yourself” approach contrasts sharply with, and indicates that he never actually applied, the Re H standard and the Lee Ming Tee principle to the respondent’s attempt to raise a case of forgery.  Nowhere is this more clearly demonstrated than in relation to his finding that the appellant was the author of the texts of Documents A and D.

420.  It will be recalled that a faint impression of the English words “One life one love” in a horizontal line, with the word “Nina” just beneath, appears on Documents A and B.  That impression was presumed to have resulted from someone writing those words on a document (called “Document E”) which was not found in the sealed envelope or before the court.  With no evidence whatsoever, Yam J purported to draw the inference that the appellant was the author of Documents D and E, with all its attendant criminal implications.

(a) First, he felt able to infer that it was the appellant who had written the missing “Document E” merely on the basis that the word “Nina” was decipherable in the impression:

“......the words ‘One life one love’ and ‘Nina’ in the form they appeared on the Documents B and A would suggest that the statement of ‘One life one love’ was probably made by a person by the name of Nina (as opposed to ‘for’ or ‘to’ Nina). Since there is no other person by the name of Nina other than the defendant herself, and in the absence of any suggestion that there could be any other person by the name of Nina who might have written the words on the paper (and it is not Nina’s case that she wrote it at the dictation or direction of Teddy), the inference and conclusion that can be drawn is that these words were probably written by the defendant herself.”  (§17.12)

(b) He used this as a stepping-stone to a finding that the appellant had written Document D with an imputation, by innuendo at least, that she was involved in forgery :

“Consequently Document D itself might well be written by her as well, when it has already been found by me both Wang and Tse’s signatures thereon are, no doubt, forged signatures. Further the fact that the draft document (‘E’) probably written by Nina did not eventually find its way into the brown envelope but the Document D with the forged signatures of Teddy and Tse did, could also support the aforesaid two inferences, i.e. Document E and in turn D were written by Nina, the defendant.” (§17.12)

421.  Yam J’s approach to the authorship of the Chinese text of Document A (agreed on all sides to have been written by someone other than Mr Wang and other than the author of Documents B and C) is equally revealing of his methodology.  Far from imposing a burden on a Re H standard on the respondent in relation to proof of the implied conspiracy (which is the only matter that authorship of such text might be relevant to), he imposed a burden on the defendant to disprove her authorship of Document A:

“......if the defendant was not the author of Document A, she must be able to know that and she must be able to prove that she did not write it. She can at least supply her own writings to her experts. (Further that could include those five English words which somehow got into Documents B and C which would cover those four words on Document D, if she wanted an expert’s opinion on them as well). It is quite surprising therefore that she had not instructed her three experts to give an opinion as to whether she was the author of Document A......” (§17.18)

422.  Later, he criticised the appellant because “she has not produced any evidence from any expert relating to the issue as to whether she was the author of any of the four documents.” (§17.40) 

423.  The enormity of this error and the sheer unfairness of this approach are obvious.  Plainly, the distorting influence of placing a burden on the appellant to “dispel all suspicious circumstances” was projected with a vengeance into the judge’s approach to the handwriting evidence.  The Court of Appeal unanimously rejected the judge’s findings as to authorship of the texts of Documents A and D.

O.6    The Court of Appeal’s approach

424.  However, except for rejecting the judge’s aforesaid findings on the supposed authorship of Documents A and D, Yeung JA essentially accepted Yam J’s approach and findings, including his finding of forgery in relation to the Tse signatures.  He found that the samples showed that Mr Wang’s signatures had “evolved” away from the style shown in the 1958 samples and rejected Professor Jia’s view to the contrary (§750), holding that similarities between the questioned signatures and the 1958 samples could not be innocently explained (§768).  His Lordship also rejected the suggestion that a calligraphic pen might have made a material difference to the signatures (§763), calling this point “yet another desperate and ill-reasoned attempt to ‘explain’ the significant differences” between the known and questioned signatures (§767).

425.  Yeung JA rejected the criticisms that have been accepted in Section O.4 above.  He held that the judge was entitled to raise matters such as the alleged retouchings without expert support since the proceedings were inquisitorial in nature (§679).  I cannot agree.  As demonstrated above, such a course was redolent with danger and unfairness. 

426.  Yeung JA also entirely rejected the argument supporting genuineness on the basis of matching variations.  He dismissed them as “very common variations in the writing of Chinese characters” which might be “accidental or natural” and in any event “entirely insignificant” (§807).  For the reasons given by Mr Justice Chan and Mr Justice Litton and developed at length by Waung J in his dissenting judgment, I consider Yeung JA to have given far too little weight to the matching variations argument. 

427.  As indicated above, Yuen JA also rejected Yam J’s findings as the appellant’s supposed authorship of the documents in question.  Furthermore, in common with Waung J, she rejected the finding of forgery in relation to the Tse signatures (although she then introduced the subtle form of forgery discussed above).  However, she agreed with Yeung JA’s finding that the Wang signatures were forgeries.

428.  Her Ladyship did so largely on the basis of her own appraisal of the questioned and known signatures.  She sought to justify this by citing Fuller v Strum at first instance and also the passage from Phipson cited above for the proposition that:

“The question of which features of the handwriting are significant and the inferences to be drawn from them are questions of judgment, assessment and opinion.” (§995)

429.  I accept that where conflicting handwriting expert views arise in the zone involving subjective questions of judgment, the view taken by the judge quite properly resolves the dispute.  However, as demonstrated above in relation to Yam J’s treatment of the optical illusion, it is plainly hazardous for a judge to assume, in the total absence of any expert’s evidence, that the issue is indeed one where the lay person’s observations are as good as the expert’s. 

430.  Thus Yuen JA thought it “obvious” that the questioned signatures resemble “more or less” the 1958 samples “but not his signatures since the 1980’s”. (§997)  I consider that a doubtful approach.  The expert literature is full of caveats against reaching conclusions on the basis of general impression.  The evidence was that factors such as the type of writing instrument, the type of paper, the speed of writing and the posture of the writer may all be influential factors affecting the general appearance of the signature, and may combine to produce a signature which may give the impression of being significantly different from known signatures written under different conditions.

431.  Yuen JA was also “struck” by the vertical layout or alignment of the characters (王德輝) as shown in a chart produced by the respondent for the purposes of cross-examining Professor Jia.  The chart in question showed an alignment of the three characters by plotting vertical and horizontal lines around them so as to frame them in a box.  The lines making up that box were positioned by first drawing two vertical lines on the plane of the strokes or dots occurring at the horizontal extremities of each signature, and then drawing another vertical line which bisects the distance between those two vertical lines.  When framed in such boxes, the questioned signatures appear to be skewed to the left of the bisecting line whereas the known signatures appear more centred.

432.  There is a certain irony in the evidence concerning the alignment of the characters.  In the somewhat discredited joint report, an attempt had been made on the appellant’s behalf to suggest that Mr Wang had a habit of writing the character (輝) slightly to the right of the vertical in relation to the other two characters in his signature and so was an indication of genuineness.  That suggestion was wholly rejected by Mr Lesnevich who described vertical alignment as “such a minor element of writing” as to be “of no value whatsoever”. 

433.  Professor Jia sought to refute the point now being taken on the respondent’s behalf by saying that the apparently skewed alignment was the result of arbitrarily choosing to plot a centre line by reference to the signature’s horizontal extremities.  He proposed several alternative ways of plotting an alignment which, he argued, took better account of the signature’s structure and which made the apparent skewing disappear.  While I can see the impressionistic appeal of the alignment argument, it must plainly be open to question whether it can be given much weight in the light of the aforesaid expert views. 

434.  Yuen JA furthermore decided that the evidence was probative of forgery on the basis of certain “unexplained significant differences” which she had identified on her own rather than on the basis of expert opinion.  Again, it is hazardous for a judge to strike out on her own in this context.  The literature stresses that for significant dissimilarities to be probative of forgery, they have to be dissimilarities which are fundamental in nature, affecting the structure of the writing.  Accordingly, the respondent’s experts were asked whether this meant that one would expect general agreement among experts as to what such fundamental differences were in a particular case.  Mr Cheng agreed that it did.  His evidence went as follows:

“Q.  But two equally competent experts examining the same questioned signatures and the same control signatures should at least be able to find the same significant differences and significant similarities? 

A.  Yes. 

Q.  Although, of course, experts could honestly come to different opinions? 

A.  As I have said, a competent expert should be able to select similar points for comparison.  They should reach similar conclusions.” (Transcript 24.76.17-77.1)

That passage in the evidence was put to Mr Tsui who agreed with the views Mr Cheng had expressed.

435.  Accordingly, doubts must arise as to the validity of the exercise where a judge identifies as “significant dissimilarities” probative of forgery certain features which were not mentioned by competent experts.  If the dissimilarities are so fundamental as to prove forgery, one must ask why they were omitted from the experts’ list.  Yuen JA’s conclusions must therefore be open to doubt.

436.  Yuen JA’s treatment of matching variations appears, with respect, to miss the point.  She dismissed the appellant’s argument in the following terms:

“As for the ‘matching variations’ exercise undertaken on behalf of the Wife, this comprised combing through the Questioned Wang signatures and the known signatures, and finding occasional common variations in writing a dot or a stroke.  It was suggested that the scattering of these ‘matching variations’ proved genuineness.  However none of these variations in the known signatures was shown to be unique to the Deceased and discoverable only by scientific methods or experts.  Nor is there any reason to assume that the person who wrote the Questioned Wang signatures had access to only some of the Deceased’s known signatures.  As such, I do not consider that there was such weight in this exercise as to counteract the effect of the unexplained significant differences.” (§1016)

437.  The probative value of a series of matching natural variations as indicia of genuineness has been discussed above.  The point is that it is exceedingly difficult to forge signatures so as to duplicate within the questioned signatures subtle and inconspicuous personal variations in the way the known signatures are signed, all the more so where a series of such matching variations can be seen.  This point does not depend on showing that the natural variations in question are “unique to the Deceased” or “discoverable only by scientific methods or experts”.  Nor is the point invalidated if one accepts that the forger may well have had access to a range of the known signatures.  However much he or she practised simulating the signature, whether based on one or several models, it would plainly be exceedingly difficult for a forger to ensure that his simulation of the signature managed to incorporate a range of inconspicuous matching variations while maintaining naturalness in the writing and avoiding other hallmarks of forgery such as hesitation, tremors or retouching.

438.  Despite Waung J’s forcefully argued and perceptive dissent on handwriting, the fundamental criticisms of Yam J in this context were not adequately addressed or corrected on appeal.  Yeung JA and Yuen JA, like Yam J, focused on matters considered probative of forgery without giving sufficient weight to countervailing considerations supportive of genuineness.  While I would not hold that the handwriting evidence impels the Court to the conclusion that the Wang signatures are genuine, it is my view that on a properly balanced consideration of the evidence for and against genuineness, the only reasonable conclusion is that the handwriting evidence was inconclusive.  It therefore does not possess much weight to be placed in the balance against the direct evidence of genuineness.

P.      Conclusions as to the evidence overall

439.  In my view, when all the evidence is taken into account, the appellant has clearly discharged her burden of proving, on the balance of probabilities, that the 1990 Will is the will of Mr Wang.  The crucial evidence in support is that contained in the statutory declaration and affirmation made by Mr Tse in which he confirmed that he had witnessed Mr Wang signing the 1990 Will and that he had signed as an attesting witness. 

440.  The fact that Mr Tse died shortly afterwards meant that he could not be subjected to cross-examination.  But the attacks on his statements proved unavailing for the respondent.  Those attacks themselves lacked credibility, having to postulate by necessary implication highly improbable multi-party conspiracy theories and having been based on faulty factual premises, or having involved unjustified inferences.  The respondent did not come near to adducing sufficient evidence to raise a prima facie case of forgery or conspiracy, applying the Re H standard and the Lee Ming Tee principle.

441.  The appellant’s own evidence in the form of her affirmations of testamentary scripts provided evidence of the will’s provenance.  However, aspects of the respondent’s attacks on the appellant’s credibility, particularly in relation to the holiday wills, diminished the weight that could be attached to her affirmations.  Moreover, the fact that the appellant’s affirmation evidence was, owing to a mistake by the respondent’s counsel, never tested in cross-examination may be thought to deprive that evidence of the weight to which unchallenged evidence would ordinarily be entitled.  The affirmations are nevertheless to be taken into account as evidence corroborating Mr Tse’s account.

442.  In finding forgery of the Wang signatures established, Yam J and the majority in the Court of Appeal applied the wrong standard of proof.  They did not, as they should have, ask whether the appellant had shown that it was more likely than not that the 1990 Will had been signed by Mr Wang.  Instead, they required her to “dispel suspicious circumstances” surrounding the will, a far more stringent standard and one based on a misapplication of the second rule in Barry v Butlin.

443.  When properly analysed, the “suspicious circumstances” identified by Yam J and those adopted by the Court of Appeal as material did not prevent the appellant from discharging her persuasive burden.  Neither was she impeded in her task by the evidence adduced with a view to questioning the propensity of Mr Wang to make a will having the characteristics of the 1990 Will.

444.  Properly approached, balancing those parts of the evidence favouring a conclusion of forgery against those parts supportive of genuineness, the only reasonable conclusion to be drawn on the handwriting evidence is that it was inconclusive and, taken with all the other relevant evidence, incapable of tipping the balance against genuineness. 

Q.      Wholesale copying and an unfair trial

445.  As previously indicated, an extraordinarily large portion of Yam J’s judgment consisted of passages copied verbatim from the written submissions of the parties without acknowledgment.  At the hearing before this Court, the appellant produced marked up copies of the judgment colouring in blue those parts copied from the appellant’s submissions and in pink those taken from the respondent’s submissions.  The vast part of the judgment rejecting the evidence given on ink-dating by Mr Speckin as a witness for the respondent was in blue.  Waung J estimated that not less than 50% of the rest of the judgment was in pink, and that in the handwriting section “the copying was something like 95% or more”.  Those estimates have not been challenged.  The appellant complains that this clearly demonstrates that Yam J did not exercise independent judicial judgment in deciding the case against her and that, distasteful though the suggestion was, at the very least, she was entitled to a re-trial.

446.  There is of course nothing wrong with a judge accepting the submissions of one party or the other where he agrees with them.  It is an everyday occurrence that a judge will adopt arguments made by one side or the other, often quoting verbatim a passage from a written argument.  Usually, of course, this is openly acknowledged with the judge saying that he accepts the submission which was put in the way set out.  However, the copying may occur to such a degree and in such a manner that serious questions may arise as to whether the judge has abdicated his judicial function or at least as to whether his conduct is such that justice has not been seen to be done by an independent judicial tribunal.

447.  This was a problem adverted to by the US Court of Appeals, Third Circuit, in Bright v Westmoreland County (380 F3d 729 (3rd Cir 2004)) where, in response to the District Court Judge’s request, attorneys for one side had submitted a proposed opinion and order of the court which the Judge had then adopted nearly verbatim as his opinion and order.  It was a particularly bad case since, as Judge Nygaard noted (at 732), the District Judge had deprived the other side of the opportunity to object or even respond to the submitted opinion and order before adopting them as his own. 

448.  The fundamental point is that a judge must bring an independent mind to his judicial function and be seen to do so.  Wholesale copying of one side’s submissions is bound to raise doubts in the mind of the other side as to whether this has occurred and whether the points made on that other party’s behalf have been adequately considered.  Judge Nygaard put it as follows:

“Judicial opinions are the core work-product of judges.  They are much more than findings of fact and conclusions of law; they constitute the logical and analytical explanations of why a judge arrived at a specific decision.  They are tangible proof to the litigants that the judge actively wrestled with their claims and arguments and made a scholarly decision based on his or her own reason and logic.  When a court adopts a party’s proposed opinion as its own, the court vitiates the vital purposes served by judicial opinions.” (at 732)

449.  When the point (referred to as the “plagiarism” point) was taken as a ground of appeal in the Court of Appeal, Waung J accepted it as a point which entitled the appellant at least to a re-trial. 

450.  Yeung JA disagreed.  While he described the extensive copying as “unconventional, surprising, and unnecessary” (§366), he pointed out that copying per se did not imply the absence of an independent judicial inquiry.  He held that Yam J had been “fully aware of and keenly interested in the issues and the parties’ arguments” (§368) and that he had “dealt with the respective cases of the parties thoroughly” (§369), having rejected some of the respondent’s arguments including on ink-dating.  Accordingly, he concluded that there had been no abdication of judicial responsibility.

451.  Yuen JA’s views were to like effect.  Having noted that the trial judge had borrowed “substantially” from the submissions, she stated:

“After a trial which overran substantially, it was probably in the hope of producing a judgment speedily that the trial judge extracted parts of the submissions for his judgment.  I have to say that this is unusual and undesirable, but I am not persuaded that it meant that no independent thought had been exercised.  It is clear from the transcript that the trial judge followed the case closely, he probed the Father’s witnesses as hard as the Wife’s, he rejected the Father’s case on ink-dating, he refused the Father’s application to amend to plead want of knowledge and approval and he made points of his own in the judgment that were not in the submissions.  I do not therefore consider that this ground of appeal has been made out.” (§902)

452.  I do not find these arguments excusing the judge convincing.  Not only was the evidence on ink-dating wholly unsatisfactory, Mr Speckin himself was wholly discredited as an expert witness for, among other things, claiming professional credentials that he lacked, claiming acceptance of his methods by the scientific community when that was false and having been trapped into demonstrating that his opinions were quite unreliable.  It would therefore have been wholly perverse for Yam J to do other than reject that evidence.  However, even then, Yam J did so by copying verbatim almost the whole of the appellant’s submissions inviting such rejection.

453.  Legitimate concerns as to whether Yam J did bring an independent mind to his judicial function do arise in the present case.  Two instances which have been examined above in some detail illustrate the grounds for such doubt. 

(a) It will be recalled that Yam J inexplicably reversed himself when, reproducing verbatim a submission to such effect, he called the appellant’s concerns causing her to resist opening of the sealed envelope “daft and illogical” whereas he had previously fully accepted those concerns as legitimate in a ruling made prior to the start of the trial (see Section K.1 above).

(b) It will also be recalled that, again copying verbatim from Mr Chan’s submissions, he once more reversed himself, going so far as to accuse Professor Jia of dishonesty when he had in the course of the hearing secured acknowledgment from Mr Chan that Professor Jia’s rejection of a suggestion would be “an end of the matter” (see Section O.4 above).

454.  The point about these examples is that they suggest that the judge reproduced the copied material without giving any genuine thought to the issue at hand.  If in each of these cases Yam J had thought independently about the issue rather than merely copying the respondent’s submissions, he would surely not have made findings wholly incompatible with considered positions he had previously taken unless he was able to articulate grounds for changing his stance. 

455.  It is therefore my view that the extent of the copying in such circumstances gives rise to doubts as to whether Yam J did bring an independent mind to bear on his judicial decision-making, this being a ground of complaint which the Court of Appeal did not sufficiently acknowledge.

456.  The point is, however, academic.  Even accepting that the appellant had legitimate grounds for doubting whether she had received a fair trial at first instance, those grounds disappeared after the 28 day hearing in the Court of Appeal.  It was accepted on both sides that the Court of Appeal was in as good a position as the trial judge to draw the necessary inferences and to come to the ultimate conclusion of fact.  There is no question of any lack of independent judicial thought in the Court of Appeal as the three separate judgments, one dissenting, make clear.  Moreover, in this Court, the complaint is academic for the additional reason that the appeal succeeds on substantive grounds.

R.      Disposal of this appeal 

457.  For the foregoing reasons, I would :

(a) allow the appeal, set aside the orders made in the courts below and order that the will of Mr Wang Teh Huei dated 12 March 1990 contained in the document marked by Yam J as “Document A” be admitted to probate as his last will;

(b) direct that the parties be at liberty to make submissions in writing as to the costs of this appeal and of the proceedings below, any such written submissions on behalf of the appellant to be lodged with the Registrar of the Court within 28 days of the handing down of this judgment; any submissions in answer to be lodged on behalf of the respondent within 28 days thereafter; and any further submissions on the appellant’s behalf in reply within 21 days thereafter;

(c) that the parties be at liberty to apply in writing for any further directions bearing on the implementation of these orders; and

(d) that any oral hearing deemed necessary to resolve any procedural issues or issues ancillary thereto arising in connection with implementation of these orders be dealt with by a single Permanent Judge.

 

Mr Justice Litton NPJ:

Introduction

458.  At trial, Mrs Wang was the defendant; Mr Wang Senior (“the father”) the plaintiff.  Mrs Wang asked the court to pronounce in favour of her husband’s last will dated 12 March 1990 in solemn form.  In reply, the father (i) denied that Document A (the will) and Documents B, C and D (the other undated testamentary papers) were made by Mr Wang and (ii) averred that the signatures of Mr Wang and Tse Ping Yim were forged. 

459.  These were the issues, purely of fact, which Yam J had to try. 

Proof of the will

460.  On Mrs Wang’s case, as set out in her affirmation of scripts, she got the documents in a sealed envelope from her husband in March 1990 : He told her it contained a will and various confidential instructions relating to the estate : It was handed to her for safe keeping and she had kept it unopened in her safe ever since, until it was lodged in court on 16 January 1998 : When her husband gave her the envelope he told her it should not be opened until after his death.  This version of events lies at the heart of the case : The manner in which the story is told is controversial, as will be apparent when the process at trial is related below.

461.  The envelope was eventually opened in court by Yam J on 6 September 1999 to reveal its contents.  Document A was the will, in Chinese, leaving all Mr Wang’s property to his wife after his death.  It is headed “will” and bears the signatures of two witnesses against the words in Chinese meaning “witness” : One signature, legibly in Chinese, is that of Mr Tse Ping Yim; the other in English signing “Lee K .... sg” has not been positively identified : It might be that of a Mr Lee Kin Sang, a contractor, who had died in June 1996.

Mr Tse Ping Yim

462.  Mr Tse was a trusted employee of the Wang family and had, by September 1999, been serving in that capacity for some 16 years.  He was then sixty-nine years of age.  After the court adjourned shortly before 6 p.m. on the afternoon of 6 September 1999, Mr Tse was contacted and, on the same evening, a solicitor Mr Alex Lam went to the offices of Chinachem in Kowloon to take his statement.  Mr Tse, obviously an intelligent and literate person, chose to write out his own statement : He went about it slowly, according to Mr Lam, taking about one hour.  Mr Lam then signed as witness at the end of the statement.  It was annexed to and became part of a statutory declaration made before the same solicitor.  (At the hearing in the Court of Appeal, it was apparently agreed that Mr Tse’s handwriting was far superior to that of the two solicitors involved in the case.)

463.  In writing out his statement, Mr Tse had copies of the four Documents A, B, C and D before him.  He circled his own signature on each of those copies and wrote “This is my signature”, then signed his name.  His statement, translated, is as follows :

“I, Tse Ping Yim, male, aged 69, HKID No. B567365(3), have been working in the household of Mr and Mrs Wang Teh Huei as a butler since 1983 and thus I have to live in the house of Mr Wang on a long term basis.

At the early stage, in about 1972 or 1973, I worked in the construction sites of Chinachem and then left Chinachem for some period of time.  In 1982 I went back to work in Chinachem again.  I have been working in the household of Mr Wang The Huei for quite a long period of time and feel that Mr and Mrs Wang treated me like their family members.  I always saw them going out and coming back together and walking hand in hand.  They were an affectionate couple.

In March 1990, Mr Wang fell from a horse.  He was first admitted to Queen Mary Hospital at about 8:00 in the evening and then transferred to St. Paul’s Hospital for treatment by neurosurgical specialists.  On that evening up until 3 am.  I was with him in the hospital all along.

He remained in the hospital for about 3 or 4 days before (he) was discharged.

Having been discharged from the hospital, Mr Wang Teh Huei immediately returned to work in his office as usual.  I accompanied Mr and Mrs Wang to go to and from the office as usual day in, day out.  Every day I would dispatch documents and deposit cheques for the company if I was free.  I vaguely remembered that after Mr Wang Teh Huei was discharged from hospital, Mrs Wang told him not to do exercise for the time being because his wound had not healed yet.

During that period of time, Mr Wang Teh Huei once asked me to go into the glass partitioned office where he worked.  At that time there were only two persons, Mr Wang Teh Huei and me, inside the room.  Mr Wang Teh Huei asked me to sign several documents already with words written thereon.  I did not read the contents of the documents because I absolutely trusted Mr Wang Teh Huei.  And, Mr Wang Teh Huei did not explain the contents thereof.  I saw Mr Wang Teh Huei signing the documents and I signed the documents too at the position designated by Mr Wang Teh Huei.  The signed documents were left on the desk of Mr Wang Teh Huei and I immediately left the office of Mr Wang Teh Huei.

On September 6, 1999 when I was about to get off work, lawyer Lam Fong Vai showed me four documents, namely Appendix 1, Appendix 2, Appendix 3 and Appendix 4, with the signatures of Tse Ping Yim thereon.  Having seen (the signatures), I confirm that all of which were affixed by me.

8:30 pm on September 6, 1999.

	(signed) Tse Ping Yim

Witness :
(signed)	Lam Fong Vai
	Lam Fong Vai, Alex”

464.  In elaboration of his statutory declaration made on 6 September Mr Tse made a second statement on 9 September, annexed to an affirmation, before another solicitor Mr Allan Leung, who took down what he said.  He repeated what he had said on 6 September and added that when he witnessed Mr Wang’s signatures on the four documents (around 6 pm that day) Mr Wang was “in a very bad mood.  He could not do any exercise because of his fall from the horse”.

465.  In his statement of 9 September Mr Tse described himself as “of Zhaoqing, Guangdong”.  On 14 September Mr Tse went to the Mainland and returned on 18 September.  On 24 September he left Hong Kong again for the Mainland.  He was admitted to Shenzhen People’s Hospital on 14 November 1999 and was diagnosed shortly thereafter as suffering from liver cancer.  He was brought back to Hong Kong on 4 December and died two days later.

466.  Pausing here, as can be seen, Mr Tse’s two statements provided powerful evidence in support of Mrs Wang’s application to have the will pronounced in solemn form.  There was no question regarding Mr Wang’s capacity to make the will in March 1990; knowledge and approval of its contents were not in doubt; Mr Tse had neither helped in its preparation nor derived any benefit from it.  He had, on Mr Alex Lam’s evidence, clearly identified his own signatures on the four documents.  Mr Tse’s statements, as they stood, constituted strong proof of the authenticity of Mr Wang’s signatures which Mr Tse said he had witnessed nine-odd years earlier.  On the face of Mrs Wang’s affirmation of scripts, (assuming it was properly before the court as evidence in the case), the envelope lodged in court on 16 January 1998 contained the last will of the deceased. 

The questions for determination

467.  The question then arises : What evidence was there before the court which could (1) refute Mrs Wang’s affirmation or diminish its weight and (2) cast doubt on the veracity of Mr Tse’s sworn testimony so that, at the end of the day and viewing the evidence (including the opinion evidence on handwriting) as a whole, the trial judge could properly conclude that due execution of the will had not been proved?  What has to be borne in mind, in the evaluation exercise, is that there is no halfway house : Either the 1990 documents are authentic, or they are forged : If the latter, then Mrs Wang and Mr Tse are necessarily parties to fraud and conspiracy : A court would not so conclude unless the evidence is compelling : The inference of serious criminal misconduct cannot properly be drawn from mere conjecture and suspicion.

468.  Unfortunately, that was not the way the matter proceeded at trial.  Almost from inception there were two matters which laid a pall of prejudice over the proceedings : They were ethereal, barely perceived at the beginning.  Ultimately they distorted the entire process and obliterated the true issues.  One of these matters was introduced by the father’s counsel, the other by the judge himself.  The two matters were these :

(a) In introducing the “main issues in the case” Mr Edward Chan SC put, as his point (1)(b), the following : “Whether the Defendant could dispel all the suspicions surrounding the surfacing and execution of the 1990 Will and her subsequent conduct in propounding the 1990 Will”. 

(b) On Day 20, during the course of the cross-examination of Mr Lesnevich (one of the document examiners called by the father), the judge said this to Mrs Wang’s counsel : He (the judge) had read the expert’s reports and asked whether counsel had considered this question : That Mr Tse’s signatures were genuine, but were put on the documents “at a later date”.  The judge went on to say that “forgery” had two meanings : “One is that it is not his signature, but a forgery can mean in a very subtle sense that it was made later on by the same person”.  Counsel immediately protested and said “that would be a totally different case” and the judge responded by saying it was for him (counsel) to consider.  The matter was not raised overtly again until the end of the trial.

Whether the defendant could dispel suspicion

469.  As to Mr Edward Chan SC’s point (1)(b) in his opening submission as referred to above, Mrs Wang’s counsel responded on Day 4 by asking why that was relevant, why it was an issue.  Mr Chan gave no answer.  The trial proceeded, lasting well over one year.  At the conclusion of the trial, the judge directed closing submissions to be made in writing, Mrs Wang’s to be served first.  These, naturally, did not deal with “suspicious circumstances” as such, which Mrs Wang needed to “dispel” in order to win.  Then, in reply, Mr Edward Chan SC set out a series of “suspicious circumstances” which, it was said, Mrs Wang had failed to “dispel” : There were ten in all, of which nine became incorporated in Yam J’s judgment, in many instances word for word.  These ranged over a far wider field than point (1)(b) in Mr Chan’s opening statement which asked whether Mrs Wang could dispel suspicion “surrounding the surfacing and execution of the 1990 Will and her subsequent conduct in propounding the … will”.

470.  In his judgment Yam J posed to himself this question : “Has the defendant discharged the burden of dispelling all the suspicions aroused in the Court that the 1990 documents were [not] the testamentary script of the deceased?”  He then examined what he considered to be the legal principles involved, citing the cases of Fuller v. Strum [2002] 2 All ER 87 (at first instance) and Hartley v. Fuld [1968] P 675.  He then said there were two questions : 

(i) “What are the suspicious circumstances which I find to be suspicious basing on those facts I found to be true”, and

(ii) “With all these suspicious circumstances whether I am satisfied that the defendant has dispelled all these suspicious circumstances hovering over the 1990 documents ...”. 

471.  Later on in the judgment under the heading “Burden of Proof” Yam J said that it was “part and parcel of the defendant’s legal burden to clear up all the relevant suspicious circumstances in light of the plaintiff’s allegation of forgery” : The judge accordingly imposed upon Mrs Wang the legal burden of “dispelling all suspicions” even though, in another passage of the judgment, he said that since it was the plaintiff who alleged forgery, it was for him to establish forgery : Not only that, but to the high standard as set out in Lord Nicholls’ judgment in Re H and Others [1996] AC 563 at 586.  It was, in the event, an empty formulation by the judge : Articulated ritualistically but never in fact applied.

Inherent bias

472.  Nowhere in his judgment did Yam J spell out what precisely was the object of the suspicion “aroused in the court”.  Underlying it all was the unexpressed premise that Mrs Wang had conspired with others (such as Mr Tse) to put forward forged testamentary papers : It was, as Waung J aptly remarked in the Court of Appeal, a case of “unalleged and unspoken conspiracy”. 

473.  This was a fundamental error which turned the trial on its head.  The judge misunderstood and misapplied the principle in Barry v. Butlin [1838] II Moore 480, as exemplified by Hartley v. Fuld and Fuller v. Strum referred to in his judgment.  In the Barry v. Butlin line of cases, the facts surrounding the execution of the will are within the knowledge of the person propounding the will; he was instrumental in preparing the will and he or someone very close to him (as in Tyrrell v. Painton [1894] P 151) takes a substantial benefit under it.  There is no injustice in these circumstances in requiring him to satisfy the court that the testator not only signed the will but knew and approved of its contents.  Knowledge and approval, in these circumstances, would not be presumed from the mere fact of signing in front of witnesses.  That was the full ambit of the principle.  It has no application if the sole issue is authenticity of the testator’s signature.

474.  Here, on Mrs Wang’s case, she knew nothing about the preparation and execution of the 1990 documents : Unless, of course, she was instrumental in their preparation.  To require her to “dispel all suspicion hovering over the 1990 documents” was to ask for the impossible.  Take for instance “suspicious circumstances” Nos. 4 and 5 together : They were, in the order appearing in the judgment, as follows : 

(i) Why did Mr Wang not ask a solicitor to prepare the will?

(ii) Why use such thin pieces of paper?

(iii) Why was it necessary to have Document D witnessed by Tse?

(iv) As Document D is in English, why did Mr Wang not sign in English?

(v) Why was the text of Document A written by one person and that of Documents B and C written by another person?

(vi) Why the internal conflict between Documents B and C?

(vii) Why the unkind expressions of sentiment in Documents B and C concerning the Wang family and Kung family?

......

(ix) Why the expression of love by Mr Wang in giving his estate to his wife?  Why was it necessary for Mr Wang to explain his reasons for so doing?

(x) Why the words “One life one love” from such an “unromantic person” as Mr Wang?

475.  Some of these might be regarded as straws in the wind which have some, perhaps slight, evidential weight on the issue of authenticity when the entire case is assessed at the end of the day.  Some might be said to weigh against Mrs Wang’s case, others could be said to be in her favour, depending on how a tribunal of fact views these matters against a somewhat hazy evidential background.  Take, for example, questions (ix) and (x) above : Mr Wang was said by some of the witnesses to be “unromantic” : Hence, such expressions of sentiment were alleged to be wholly out of character and provided circumstantial evidence of forgery.  But there was also strong evidence before the trial judge to the effect that for some twenty years before March 1990 Mr and Mrs Wang were, as far as people who knew them could judge, a close and loving couple, living together and working together to build the Chinachem Group into a huge financial empire.  So, it might be asked rhetorically : Why would Mr Wang not leave some expression of love for his wife in depositing his will with her?  However the trial court might choose to view these peripheral matters – and a tribunal of fact has some latitude in this regard – there must be balance and fairness in the exercise.  They are peripheral pieces of evidence in the case : No more : To be weighed in the scales with the evidence as a whole.  To cast upon Mrs Wang the legal burden of dispelling “suspicion” arising from these factors is fundamentally wrong in principle. 

476.  As mentioned above, some of this material points both ways, on the only issue that counts in the case : forgery.  Take another example : Question (v) above : How does suspicion of forgery arise unless Mrs Wang was in some way involved in wrongdoing?  If matters were slanted the other way, one might equally ask : Why would a forger have involved more persons than necessary in such a risky criminal enterprise?  Or take the unkind expressions of sentiment in Documents B and C : Would a forger have chosen to provoke controversy and resentment in that way?  These are highly speculative matters and gain weight only if the scales are tipped one way or the other by the tribunal of fact.

477.  The case from the start was slanted against the defendant.  With regard to the unkind expressions of sentiment in Documents B and C, the judge said that they “could not be the expressions of Teddy himself” (copying verbatim from Mr Chan’s written submission).  He then went on to say :

“In conclusion, if one has to choose between Teddy and Nina, I think the expressions that the Wangs family members were disappointing and the Kungs family was disgusting suited the sentiments of Nina much better than Teddy.  Could it be an expression of Nina instead of Teddy in these questioned documents?  Could this be a case of ‘projection’ of Nina in psychology onto Teddy which was eventually found in Teddy’s purported ‘will’?  Would Nina add those words into the will also for the purposes of making it more convenient for her to refuse the Wangs and the Kungs when they should ask her for any payment from the enormous estate of Teddy?” 

478.  The judge then referred to the “clear evidence” of Mr Yih and Teresa Wang – who testified as to events more than 20 years before March 1990 – and said that it was “out of character” for Mr Wang to use words like “One life one love” in Document D, and likewise the words in Document B “I love my wife.  After my death, all of my property, real estate and my body shall belong to my beloved wife” : These were, he said, “probably not the expression of Teddy”. 

479.  Having reached this point, the game was over : The judge had in effect decided the case against Mrs Wang.  It is not therefore surprising that he went on to find as a fact that Mrs Wang herself was the author of Documents A and D : This was unanimously set aside by the Court of Appeal.  Yeung JA said he was surprised at the judge’s conclusion on Document D : It “defied comprehension” : This is the sad consequence of a trial judge allowing peripheral issues to spread like cancerous cells, subverting the discipline of law. 

“Subtle form of forgery”

480.  At the conclusion of the trial, the judge found the four Tse signatures forged : Therefore the question of a “subtle form of forgery” which he had raised on Day 20 – that Tse had assisted in the preparation of false documents by signing at a much later date – became no longer relevant.  But, as will be seen later, this crept back into the case in the Court of Appeal, with unfortunate results. 

The Court of Appeal

481.  In the Court of Appeal Waung J correctly held that the Barry v. Butlin line of cases was relevant only where the issue was knowledge and approval : The principle was inapplicable when the sole issue was authenticity.  He conducted a painstaking analysis of all the evidence and, in reversing the judge, would have pronounced in favour of the will in solemn form.

482.  The Court of Appeal was unanimously of the view that the “suspicious circumstances” standing alone were not enough to sustain a finding of forgery.  But this was, unfortunately, where unanimity ended.  Leaving the dissenting judgment of Waung J to one side, we find the following :

(1) Yeung JA began his analysis of the issues by saying :

“... the wife had no burden to disprove the allegation of forgery based on the ‘suspicious circumstances’, even if they were expressly pleaded (which they were not).  In fact, it was the Father who had the burden to establish forgery with strong and cogent evidence”.

It might be thought, therefore, that the learned Justice of Appeal was saying that the principle in Barry v. Butlin was inapplicable and that the trial judge had erred in this regard.  But, almost in the same breath, Yeung JA said that Mrs Wang had to prove due execution of the will and therefore “might have to dispel the ‘suspicious circumstances’ if such ‘suspicious circumstances’ were inconsistent with the due execution of the 1990 documents” : Importing, in effect, the same perspective into his own analysis of the circumstantial evidence as that adopted by the judge.  This will be dealt with in greater detail below.

(2) Yuen JA began her judgment by saying :

“It is clear that the legal onus of proving a will is on the party propounding it, in this case, the Wife.  It is also indisputable that as the Father was alleging that the signatures of the Deceased and Tse were forged, he had the evidential burden of proving those two allegations … The standard required of the Father, to prove that the signatures of the Deceased and Tse were forged, is … on a balance of probabilities, but by reason of the seriousness of the allegations, ‘the more improbable the event, the stronger must be the evidence that it did occur before, on the balance of probabilities, its occurrence will be established” (Re H [1996] AC 563, 586-7)”.

The learned Justice of Appeal went on to say that “the body of case law relating to ‘suspicious circumstances’ in the context of want of knowledge and approval” was not relevant : Thereby overturning the trial judge’s approach to the evidence : Rightly so.  How, then, did she come to dismiss the appeal?

Unfortunately, the learned Justice of Appeal viewed forgery and due execution as two separate issues : They are not, she said “opposite sides of the same coin … even if the evidence is not strong enough for the Father to prove forgery, the evidence he has adduced towards that end, together with other evidence, may still instill such doubts in the court’s mind that the court finds that it cannot grant probate of the will being propounded” (emphasis added).

The doubts referred to in her judgment are those which Mrs Wang had failed to dispel : Doubts which led the trial judge and Yeung JA to conclude that forgery and fraud had been proved : Thereby, in her own review of the facts, admitting through the backdoor Barry v. Butlin which she had barred at the front. 

(3) Yuen JA, agreed with Waung J on a crucial issue : The judge’s finding that Mr Lam the solicitor had lied concerning the circumstances under which Mr Tse had written out his own statement on 6 September 1999 could not stand : The statement had not been “pre-prepared” as Yam J had found : Hence the truth regarding how the statement had come into being is as related in para.462 above.

(4) Yuen JA agreed with Waung J that the finding of forgery regarding Tse’s signatures could not stand.  But then she added :

“I should however clarify that this should not be taken as an acceptance of the submission that Tse’s signatures on the documents being propounded were made in 1990”.

With this one sentence, the learned Justice of Appeal left the suspicion of “a subtle form of forgery” hanging in the air over Mrs Wang.

483.  Adopting the approach they did, it is perhaps not surprising that the majority found many matters unexplained by Mrs Wang.  For example :

– Why did Mr Wang, who had taken the precaution of engaging solicitors for his two previous wills, leave the preparation of Document A to such a seemingly untutored scribe?

– Why did Mr Wang say in Document B that his parents disappointed him when he was apparently close to his mother and was on good terms with his father?

Asked in a neutral tone, these questions lead nowhere on the issue of execution : Put on the basis that these were matters Mrs Wang had to explain, the implication was that the purported testator had not signed those documents and had not approved their contents; the further implication was that Mrs Wang was a party to the fraud : These matters constituted, according to Yam J, a “projection of psychology” by Mrs Wang onto the fictitious persona of the deceased.  Regrettably, the majority of the Court of Appeal adopted this warped approach.

484.  Before leaving the judgments of the Court of Appeal there are six matters I must deal with : These have some bearing on the issue of forgery, of greater or lesser weight :

(1) The evidential effect of Mr Tse’s statements.

(2) Mrs Wang’s affirmation dealing with the provenance of the will.

(3) The other attesting signature “Lee K ... sg”.

(4) Mrs Wang’s behaviour regarding the envelope.

(5) Mrs Wang’s failure to testify in Court.

(6) The opinion evidence of the experts.

The weight of Mr Tse’s statements

485.  Yeung JA said that it was “inconceivable” that Tse “without prompting or coaching of any kind” could have come up with statements which provided a “perfect match with the Wife’s case”; Yuen JA said that it was remarkable that Tse should have “volunteered information” concerning Mr and Mrs Wang acting in a loving way towards each other.  She said :

“There was no evidence which explained why Tse would have taken it upon himself to volunteer the passage when he was writing the first statement”.

486.  This, with respect, is again to view the matter in a slanted way.  Mr Tse’s statements were made in September 1999, after the envelope was opened by Yam J in court and the contents revealed.  Long before then, in the course of bitter litigation for many months, the father had made accusations of criminal misconduct against Mrs Wang : That the envelope and its contents had been “made up” by her in order to thwart his attempt to take control of the Chinachem Group of companies.  This had generated much publicity in the popular press.  For example in the 22 January 1999 issue of Next Magazine (with photographs of the father and Mrs Wang prominently displayed on the front cover) there was this statement :

“[Mrs Wang’s] lawyer said in court that Wang Teh Huei gave Nina a sealed envelope with a will and other things two months before he was kidnapped in 1990.  She does not know the contents of the will as she has not opened the envelope.  Wang Din Shin does not believe her claim.”

487.  It would be reasonable to infer that Mr Tse was aware of these accusations : He was necessarily implicated in the accusations of criminal misconduct.  So why should he not provide as full a statement as possible when given the opportunity?

488.  Then it was said that Mr Tse had “made himself unavailable to the Hong Kong police” when he left Hong Kong for the Mainland on 24 September 1999 : This was after the judge had given the father leave to report the allegation of forgery to the police on 23 September.  Yuen JA said that Mr Tse, by leaving for the Mainland, “never allowed himself to be questioned by an independent investigator”.  The unspoken accusation is that he had something to hide.

489.  The learned Justice of Appeal’s approach is unsound.  There was no evidence that Mr Tse knew of the order of 23 September.  He had in fact applied for a visa to go to Thailand the previous day.  He had left Hong Kong once before in the same month : On 14 September for the Mainland, returning on 18 September.  He was, it appears, a frequent visitor to the Mainland where he was having a house built.  He had not been approached by the Hong Kong police before he left on 23 September.  But for the fact that he fell ill in November and died on 6 December he would, presumably, have been available for interview by the police sooner or later.  Nothing suggests that he intended to stay away permanently.  It is difficult to see how his departure on 24 September reflected upon the reliability of his statements : Unless, as Waung J remarked, the case advanced by the father was one of “unalleged and unspoken conspiracy”.

Mrs Wang’s affirmation of testamentary scripts

490.  If doubt surrounds the precise status of Mrs Wang’s affirmation of testamentary scripts at trial – whether it was regarded as evidence placed before the trial judge – there was none when the matter went before the Court of Appeal : Counsel for the father accepted that the court could refer to it as evidence in the case.  Before us Mr Neville Thomas QC, counsel for the father, made the same concession, adding that “it falls to be considered whether it is admissible for its full evidential weight”.  This brings up a dilemma.  On the one hand it could be said : “Here is sworn evidence before the court, unchallenged, that sometime in March 1990 Mr Wang gave his wife an envelope containing the four documents; it could hardly be supposed that Mr Wang would have handed to his wife forged documents; the matter is conclusive in her favour.”  On the other hand it might be maintained : “No formal order was made under O.38 r.2 of the Rules of the High Court allowing affidavit evidence to be given, so the court must treat the position as if there was a total void regarding the provenance of the four documents”.  Neither position is satisfactory.

491.  The dilemma arises partly from the unsatisfactory way the trial was conducted.  Plainly, Yam J treated the position as if there was before him affidavit evidence from Mrs Wang, untested by cross-examination; its evidential weight, in the eyes of the judge, became totally submerged by the “suspicious circumstances” from which the inference of fraud and conspiracy was drawn : So, as far as the trial judge was concerned, Mrs Wang’s story on affidavit must be untrue.  In the Court of Appeal, Yeung JA adopted essentially the same stance : He said, for example, that it was “inconceivable for Wang to have executed a will (one concerning assets worth billions of dollars) by way of the 1990 documents and without the help of lawyers as he did for his two previous wills”; he went on to hold that it was comical to suggest that Mr Wang should make a will in 1990 in favour of his wife, with the declaration “One life one love”, whilst keeping the 1968 will in his father’s favour together with photographs of his ex-girl friends in the safe deposit box : So, plainly, as far as Yeung JA was concerned, Mrs Wang’s story on affidavit was untrue.  He never took the position that the affirmation should be disregarded as evidence before the court : He simply took issue with counsel’s suggestion that the affirmations “should be given full weight because the father did not apply to cross-examine her on her affirmations” : He described it as “a rather astonishing submission”.

492.  Yuen JA’s view of the material before the court was not as extreme as that taken by Yeung JA : She considered it highly significant that Mrs Wang, the only “living link to the res gestae”, chose not to testify in court : Although Yuen JA commented on the fact that no order had been made by the trial judge for Mrs Wang’s affirmation to be used in place of oral evidence, she did not take the stance that the affirmation must be wholly disregarded.

493.  Unsatisfactory though the position is, relating to a matter which has a direct bearing on the issue of authenticity – namely, the provenance of the will – the only proper stance to adopt is, in my judgment, that suggested by Ribeiro PJ in the course of argument : That there is some evidence of provenance before the court, which might be undermined or rebutted by other evidence or relevant factors, such as the way Mrs Wang behaved in the course of the proceedings : Essentially the same stance adopted by the Court of Appeal.

Lee Kin Sang

494.  In my judgment, far too much was made in the lower courts (Waung J excepted) of the fact that the appellant’s lawyers never pinned down the precise identity of the other attesting witness who had signed as “Lee K… sg” on Document A.  If it was the signature of one Lee Kin Sang, which seems possible, he was a registered contractor employed as chief supervisor of Chinachem’s building sites, and had an office one floor below that of Mr Wang.  He had died in June 1996.

495.  No allegation was made in the respondent’s pleadings that the signature “Lee K… sg” appearing on Document A was forged : So neither party led opinion evidence from experts on that subject.  Given the way the trial evolved before Yam J, it would probably not have helped the appellant’s case if genuineness of that signature were indeed established on the basis of expert evidence.  The riposte would simply have been : “You have not shown when Lee Kin Sang signed; he could have signed any time before he died in June 1996” : Short of calling Lee Kin Sang to testify in court, it would not have been possible for the appellant to overcome the hidden premise underlying the whole case : That she was a party to a conspiracy to put forward forged documents and to pervert the course of justice.

496.  As to Mrs Wang’s failure to pin down the identity of the person signing – or to explain what steps were taken in that regard – it was equally open to the respondent to do that : And, if one were to slant the analysis against the respondent, it might equally be said : The respondent was reluctant to take such a step because he knew that a positive identification would have virtually concluded the case in Mrs Wang’s favour : If the attesting witness turned out indeed to have been Lee Kin Sang, he could well have been on hand to witness Mr Wang’s signature on 12 March 1990 since he had his office just one floor below that of Mr Wang.  Surmise and speculation provide no firm base from which to draw inferences : Particularly in relation to such serious matters as fraud and conspiracy.

Mrs Wang’s behaviour regarding the envelope

497.  The majority of the Court of Appeal considered Mrs Wang’s “reluctance” to produce the will significant : This is allied to Yeung JA’s remark that it was “very strange and suspicious” that the existence of the envelope was never made known to the police investigating the kidnapping.

498.  Mrs Wang’s position has been misunderstood.  Her relationship with the respondent was, up to March 1997, cordial.  The respondent had, for years after the kidnapping, attended Chinachem’s offices daily to read newspapers.  Then, in April 1997, soon after the expiration of seven years from the date of Mr Wang’s kidnapping, out of the blue the respondent launched proceedings for leave to swear death.  Mrs Wang opposed the application on the basis that she believed that Mr Wang was still alive : She had paid a huge sum on 26 April 1990 as part of the ransom demanded and had been contacted by the kidnappers from time to time since then : She was also in contact with the Hong Kong and Taiwan police concerning the kidnapping : Their investigations were for kidnapping, not murder : Her position was that so long as her husband was alive she should not open the envelope and that to do so might jeopardise her husband’s safety.  That was her position throughout those earlier proceedings (Probate Action No. 6 of 1997).  Yam J in his judgment dated 29 April 1998 accepted Mrs Wang’s evidence that her husband had told her not to open the envelope until after his death, remarking that otherwise there might be “irreconcilable consequences” and that her fears as to her husband’s safety were “not illusionary”.

499.  Nothing had changed between Yam J’s judgment in Probate Action No. 6 of 1997 and the trial in this case : And yet we find, in Yam J’s judgment the following :

“The defendant had shown extreme reluctance to disclose to the plaintiff that she was in possession of the 1990 Documents …  There were really no reason for the delay in … filing … the affirmation of Testamentary Scripts in Probate Action No. 6 of 1997 …  Eventually under the compulsion of the ‘unless order’, after the defendant had filed her first affirmation of Testamentary Scripts, she still tried every means of preventing its contents from being revealed … because of two purported reasons :

(a) concern about the safety of Teddy; and

(b) respect of the wish of Teddy who told her that he did not wish the testamentary arrangement to be revealed in his lifetime.

Both reasons are daft and illogical.”

500.  This passage was copied almost verbatim from Mr Edward Chan SC’s written submission : Including the adjective “daft”.  It stands in stark contrast with the judge’s earlier finding.  The only logical explanation for it is that he had blindly copied counsel’s submission and had made no independent evaluation of the point at all.  Far from overruling the judge in this regard, Yeung JA said that Mrs Wang’s explanation was “highly unconvincing if not outright ridiculous”.  The basis for his strictures was not explained : Underlying them is the “suspicion” that Mrs Wang had fabricated the 1990 documents long after March 1990 : A “subtle form of forgery”.

Mrs Wang’s failure to testify in open court

501.  In the normal course of events one would have expected the person who had received the testamentary papers from the hands of the testator to testify in open court regarding their provenance.  In the circumstances of a case such as this, if believed, it would have gone a long way to prove due execution of the will.  It is clear that when the case was opened before Yam J the expectation was that Mrs Wang would be called as a witness.  But this was no normal trial.  The judge exercised no control over the proceedings and allowed the evidence to roam over a scandalously wide field.  Three witnesses for the respondent, the father himself, Mr Yih Lee Kong and Teresa Wang, testified for days on end regarding matters in the remote past, prior to 1970 : Scandalous allegations of adultery were allowed to be made against Mrs Wang back in 1968 : Mr Yih was cross-examined by Mr Martin Lee SC for 20 days and the father for 7 days; much of it was, according to the judge, “irresponsible, vicious, opportunistic” and “just a waste of time”.  But he did nothing to stop it.  Yeung JA said that the “aggressiveness, oppressiveness, unfairness and aimlessness of the cross-examinations were exceptional”.  Yuen JA said that “the evidence on virtually every point, as well as every non-point, was pursued to extremes on both sides”.

502.  On Day 151 of the trial, when the case had grossly over-run and the judge was discussing with counsel its further progress, the question of the time required for Mrs Wang’s testimony arose.  Mr Edward Chan SC, counsel for the respondent said this : If Mrs Wang be called, her cross-examination would be long : whether as long as Mr Yih’s, counsel was unable to say, but he “would not be surprised [if] the defendant should hold the record, after all”.  The only reaction this drew from the judge was for him to say, “in light of the evidence given by the [the] various witnesses” counsel should take a further statement from Mrs Wang.  The “various witnesses” mentioned by the judge included, presumably, Mr Yih, Teresa Wang and the respondent himself, speaking of matters before World War Two and the Wangs’ marital difficulties in the 1960s.  There was no hint that the judge was likely to circumscribe the scope of the cross-examination or to protect the witness from oppressive questioning on scandalous and irrelevant matters.

503.  This is the context in which the question of failure to testify must be viewed.

504.  Tied up with this question is the earlier one : The precise status of Mrs Wang’s affirmation placed before the trial court.  It was never suggested at trial that the judge should proceed on the basis of a complete lacuna in the evidence, that the testamentary papers simply evolved out of thin air : Rather the position was taken that Mrs Wang had chosen not to have her story tested by cross-examination.

505.  Prior to Day 169, when Mr Martin Lee SC told the judge that he was not calling Mrs Wang as a witness, there had been a number of witnesses giving evidence concerning the closeness of relationship between Mr and Mrs Wang, and her own contribution to the success of Chinachem : Chinachem was, according to some of the witnesses “their baby”.  There was, of course, the evidence of Professor Jia on the genuineness of the signatures on the will.

506.  The position might well have been taken by those advising Mrs Wang that there was, before the court, already sufficient evidence to establish due execution of the will : There was little she could add to the case by way of evidence.  On the other hand, as Waung J remarked in his judgment, if driven to give evidence, Mrs Wang’s whole private life would have been explored in the greatest graphic detail with all the attendant publicity : It would have been obvious to Mrs Wang’s advisers much earlier than Day 169 that she could expect no protection against oppressive cross-examination from the judge : A point which Mr Neville Thomas QC countered by saying that she could have sought an adjournment of the trial and put the matter before the Court of Appeal on an interlocutory appeal : Hardly a satisfactory solution.

507.  Viewed in the light of the matters as set out above, Mrs Wang’s failure to testify carries, in my judgment, far less weight than was thought by the majority of the Court of Appeal.

Handwriting

508.  In my judgment, Yam J’s findings on forgery were fundamentally flawed, as will be explained below.

509.  With regard to Mr Tse’s signatures the judge’s finding has been over-turned by a majority in the Court of Appeal.  Mr Neville Thomas QC accepts that, in this Court, we should proceed on the basis that Tse’s signatures are genuine : This demonstrates the fragility of the opinion evidence adduced in support of the father’s case and immediately leaves a question mark over the conclusion that the Wang signatures are forged : Particularly as two of the three experts called by the father were far less certain of forgery of the Wang signatures than they were concerning the Tse signatures.  Furthermore, as Mr Patrick Cheng Yau Sang, one of the father’s three experts, said : Two equally competent experts examining the same questioned signatures and the same samples should be able to find the same significant differences and similarities : There was surprising divergence between the father’s three experts in this case.

510.  The expert called to testify by Mrs Wang was Professor Jia Yu-wen, a professor of the Criminal Police College of China and a staff member of the Material Evidence Appraisal Centre of the People’s University of China.  The Centre was approached sometime in the year 2000 by Mrs Wang’s solicitors to conduct an appraisal and give a report on the eight suspected Wang and Tse signatures.  Three staff members were assigned to the task : Professor Xu Li-gen, Professor Zhan Chu-cai and Professor Jia.  They were originally provided with five samples of Mr Wang’s signatures, on documents dated 1984 and 1985.  In October 2000 they came to Hong Kong to work in the Government Laboratory and were given more samples.  Eventually, when Professor Jia testified before Yam J he was working on 18 samples, of varying quality.  Most of these were made with ball-pens : Not ideal as samples as the questioned signatures were made with a calligraphic pen.

511.  As regards the structure and general characteristics of the Wang signatures, the conclusion of the appellant’s experts was that they were free-flowing, the stroke movements were natural, the degree and pace of the pen pressure rhythmic and the paper was scratched through and cut in placed due to the pen pressure : All highly probative of genuineness.  There was no real challenge from the father’s experts on these points : In fact, one of the father’s experts Mr Gus Lesnevich specifically agreed with these observations.

512.  By the use of a stereo microscope and an infra-red instrument Professor Jia and his colleagues found also the following : There were no features of imitation in the four Wang signatures such as pauses, rewriting of strokes, rigidity or awkwardness; the strokes were slowly executed.  Mr Lesnevich agreed with all these points.

513.  Professor Jia and his two colleagues made their written report on 8 December 2000.

514.  When Professor Jia eventually testified in court, he took 17 days in examination-in-chief.  Part of that time was spent examining signatures under a microscope, with enlarged images projected onto a screen.  Obviously, his testimony in court was far more detailed than the contents of the joint report.  Two matters which reinforced his opinion as to authenticity were never mentioned in the joint report : (i) Matching variations (as to which elaboration will come later) and (ii) the significance of the use of the calligraphic pen in executing the suspect signatures.

Matching variations

515.  Each character in a Chinese name is made up of separate strokes of the pen.  Sometimes the strokes are linked, sometimes not.  For example, the character Huei (輝) is made up of 15 strokes.  If a person writes that character four times, the pen strokes might be identical, or there might be variations in the way a particular stroke, or a combination of strokes, is executed.  These variations, in some cases subtle and by no means obvious, are peculiarities of the writer.  If one finds the same peculiarities in the sample signatures – that is to say, they match – then these matching variations become strong evidence of genuineness.  The more the matching variations, the stronger the inference becomes.  That is because the forger, carrying one mental design in his head when he simulates the genuine signature, is unlikely to be able to simulate all the variations : It would be an extraordinarily skilful forger who would be able to carry the range of variations found in the genuine signatures in his head and to put into his simulated signatures the same variations.

516.  What is set out above is plain common sense, supported by the leading textbooks on the subject of suspect handwriting and agreed in principle by the respondent’s three experts : For example, Mr Patrick Cheng Yau Sang agreed with this proposition put to him in cross-examination :

“The forger does not understand this necessity for natural variations and, as nearly as he can, makes words and letters just alike.”

517.  In relation to the four Wang signatures, Professor Jia found nine matching variations : Obviously of varying degrees of relevance.  One example would suffice.

518.  In the character Teh (德) the first two strokes of the “double men” radical (彳) would normally be executed from right to left.  However, in two of the four suspect signatures Professor Jia found on microscopic examination that the first of the two strokes was written the other way round, from left to right; he said it was almost impossible for one to take note of this.  Professor Jia was not challenged on this aspect of his evidence.

519.  In relation to a number of samples, other variations appear : They match with the suspect signatures : Not all of these can be said to be peculiar to the writer, but on any view of the case, a number of matching variations would provide some evidence of genuineness.  This then has to be assessed with all the other matters found in order to make, in the end, a global assessment of genuineness or forgery.

520.  Neither the narrow point concerning the first stroke of the “double men” radical (彳) nor any of the other points of matching variation was addressed by Yam J.

521.  In relation to the way the “double men” radical (彳) was executed in Document D (the exhibit used was Ex. 1D1) the judge compared that first stroke – executed from left to right – with sample Ex. No. DPH 668 where it was the other way round.  This, as any one can see, is a point of difference.  And if one focused simply upon this difference between Ex. 1D1 and Ex. No. DPH668 of course one would say that this was a dissimilarity suggestive of forgery : Thereby missing, as the judge did, the entire point Professor Jia was seeking to make.

522.  The point is simple, and bears repetition.  It can be shown from the specimen signatures that Mr Wang had two ways of writing the first stroke of the “double men” radical (彳) in the character “Teh” (德) : Ex. DPH668 was one way, Ex. DPH669 was another : Internally, between Ex. DPH668 and 669 there was a variation : In Ex. DPH669 that stroke was written from left to right.  And the same variation can be found internally among the four suspect signatures.  Of course the forger could have picked up the subtle variation when he was studying Mr Wang’s known signatures before he embarked upon his (or her) task of simulation and embodied that variation in the forged signatures : But on any view of the case this was evidence pointing towards genuineness.  And, as mentioned earlier, the more the matching variations, the greater the probability of genuineness.

523.  The judge turned Professor Jia’s point on its head and held him guilty of “intellectual dishonesty” : He said that Professor Jia eventually conceded that the “double men” radicals (彳) seen in Ex. 1D1 and Ex. DPH668 were different.  The answer is : Of course they were different : Mr Wang, according to Professor Jia, had two different ways of executing that radical in his signature.  Professor Jia could not have made the point clearer when he said (Day 102 p.68-69) : “You just look at 668 and 669 … they are written in different ways … would you say that these two were in fact not written by Mr Wang?”  The judge never took that point on board and found him guilty of intellectual dishonesty on a specious comparison.

524.  The passage in which the judge accused Professor Jia of intellectual dishonesty was copied almost verbatim from Mr Edward Chan SC’s closing submission : It is regrettable that such a charge should have been made by counsel : It is even more regrettable that Yam J should have embodied it in his judgment : In the Court of Appeal only Waung J sought to correct this injustice.

525.  As Waung J highlights in his judgment, Mrs Wang’s counsel had, in his written submissions, made the following point (said to be the single most important point in the case) : There were at least 9 significant variations found among the 18 specimen Wang signatures : These were reflected in different ways in one or other of the four questioned signatures : It would have taken “the devil incarnate” to have reproduced these variations in the alleged forgeries.  This point was simply ignored in Yam J’s judgment : It was inconsistent with the conclusion he had already formed that the signatures were forgeries (all the signatures, including Tse’s ).  Given the way the trial proceeded, how could it have been otherwise?  In relation to the “suspicious circumstances” the judge had said :

“Those suspicious circumstances together can only point to one irresistible conclusion and that is the 1990 Documents are nothing but forged documents.”

And then, to knock another nail in the coffin of the defendant’s case, he had added :

“If the burden is on the plaintiff to prove the 1990 Documents were forged documents as he has alleged forgery in his pleadings, I must say that he has discharged this burden by the evidence presented before me.  All these pieces of evidence concerning suspicious circumstances alone are strong and cogent evidence.  The only irresistible conclusion that one can draw from all the unanswered or unanswerable, or unexplained or unexplainable questions raised by the evidence presented by the plaintiff, is that the 1990 Documents are forged documents.”

526.  With this conclusion in mind, how could the judge have evaluated the handwriting evidence fairly or entertained doubts arising from such evidence?  There is, in my judgment, considerable force in Waung J’s observation that the trial judge had “created a built-in loading in favour of forgery and against genuineness and this vice poisoned the whole of the handwriting analysis …”.

527.  Nowhere is this “vice” more apparent than in the question of retouching, to which I now turn.

Retouching

528.  Obviously, where the writer of a suspect signature is found to have retouched or embellished some strokes, that would mean that those strokes were not free-flowing and natural : This could indicate forgery.

529.  Here, none of the three experts called by the plaintiff found signs of retouching.  In re-examination Mr Lesnevich was asked whether there were features such as retouching, tremors and things of that sort : He replied :

“Certainly not in this case.”

530.  And yet the judge, on his own, found nine instances of retouching from which he concluded that they were all signs of unnatural writing and forgery.

531.  In reaching these conclusions the judge was working from photographs and images of the magnified signatures captured on the computer.  With regard to the shading where two thin lines intersect (thin, that is, in the original signatures), the barely perceptible difference in shading as shown in a photographic enlargement may not accurately reflect the reality in the original.  This the judge ignored.

532.  Further, in the standard textbooks on suspect documents, there is discussion of a “well-known optical illusion” : Where two lines intersect, the darker invariably appears to have been written over the lighter, irrespective of the true state of affairs : see “Suspect Documents” by Wilson R. Harrison (Sweet & Maxwell, 1966) p.226 and “Questioned Documents” by Albert S. Osborn (Sweet & Maxwell, 1929) p.510.

533.  Take one of the instances of the judge finding retouching in the Wang signatures where the experts found none : In the vertical stroke in the Wang character (王) where it intersects the middle horizontal stroke.  Professor Jia said in evidence that the ink was heavy and thick for the vertical stroke, and when the pen turned to make the horizontal stroke the “pen wielding movement” was faster and the pressure lighter : Hence the ink left on the paper for the horizontal stroke was more faint.  When counsel for the father put to Professor Jia the suggestion that the downward vertical stroke had been written over the horizontal stroke, Professor Jia disagreed.  That should have been the end of the matter.  There was nothing to contradict his conclusion.

534.  And yet we find, in counsel’s final submissions on this point, a reference to Professor Jia’s purported explanation and the assertion – without the least evidence in support – that there was “some retouching of the downward stroke”.  The judge simply copied those submissions verbatim and made them part of his judgment.

535.  It would needlessly burden this judgment to cite more instances where Yam J’s judgment reproduced counsel’s submissions on retouching as if they were his own findings, and in the process categorized Professor Jia as dishonest.  It is regrettable such submissions were ever made.  If what the judge said by way of wholesale copying from counsel could be categorized as any sort of finding I would hold that his findings of dishonesty are unsupportable.  They are grossly unfair to Professor Jia.

536.  The findings of retouching by the judge were central to his conclusion on forgery.  There was no evidential support for his finding.  And yet they were uncorrected in the Court of Appeal (Waung J excepted).

Calligraphic pen

537.  The following points are borne out by the evidence : (1) The signatures were executed with a broad-nib calligraphic pen; (2) they have marked similarities with the 1958 samples DPH668 and DPH669 which were done with a fountain pen; (3) among the 18 samples, only one other (in 1980) was made with a fountain pen; (4) signatures made with a calligraphic pen would tend to be slower and more deliberate than done with a ball point pen; (5) Mr Wang had more than one design of his Chinese signature.

538.  The type of pen used is highly relevant when making comparisons : This is clearly borne out in Professor Jia’s evidence, as to which there was no challenge.  This explains, to some extent, the puzzle in the case : Why did the four signatures resemble so closely the 1958 signatures and not the later signatures?  The answer might lie in the fact that the 1990 signatures appear to be made more slowly, deliberately, with a broad-nib pen; the post-1967 signatures were more speedily executed and were made with ball-point pens.

539.  It logically follows from this that the similarities of the suspect signatures with DPH668 and 669 provide at least some evidence of genuineness : Unless one assumed (i) that Mr Wang’s signature had deteriorated over the years since 1958, as Mr Lesnevich did, or (ii) that Mr Wang had discarded the pattern of his signature as seen in DPH668 and 669 since that time, as Yeung and Yuen JJA did.  Yeung JA spoke of the “evolution” of Mr Wang’s signature over the years.  But none of these assumptions was justified on the evidence.

540.  In short, evidence of authenticity relied on by Professor Jia, based upon the characteristics of the calligraphic pen, and the more deliberate pen movement resulting from its use, was simply disregarded in the courts below : There is no mention whatever of the type of pen used to execute the signatures on the four documents in Yam J’s judgment.  Moreover, the judge ignored evidence from the father’s own experts which tended to explain some of the perceived differences between the suspect signatures and the samples.  For instance in para.10.55 (6) of his judgment Yam J referred to the middle loop in the Wang character (王) which, in three of the four suspect signatures, was “virtually closed” : In the same character executed with a ball-point pen, the loop is open : The judge considered this a significant difference : There was in fact evidence from Mr Tsui to the effect that it was more difficult to produce an open space with a broad nib calligraphic pen than with a ball point pen : Which seems commonsense.  This the judge ignored.

Conclusion on Handwriting : Trial Court

541.  The approach of the judge, as Waung J rightly pointed out, was to look for apparent disparity in isolation, adopting (and copying) the written submissions of Mr Edward Chan SC.  It was an adding up by numbers.  He never conducted a consolidated evaluation of the signatures, taking into account all relevant factors including the pen type, pen pressure, ink infusion and the speed of execution.  It has also to be remembered that the available specimens of Mr Wang’s Chinese signatures were few : Only 18, spanning the years 1958 to 1985.  There were many variations in the way the same strokes were executed within the 18 known signatures.  If one were to take DPH668 as a suspect signature and compare that with the other 17 then, adopting Yam J’s approach, DPH668 would be found to be a forgery : There would be many points of dissimilarity.  Any significant similarities would be treated as evidence of simulation.

542.  There were, in the four Wang signatures, minor signs of tremor which the father’s experts considered of no significance.  In the joint report produced by Professors Xu, Zhan and Jia, they linked the tremors to the injuries sustained by Mr Wang when he fell from his horse on 10 March 1990.  Whether this was going too far (as Yeung JA thought) is debatable : Perhaps all that can properly be said is that the tremors are consistent with the fall : What is unacceptable is the judge’s conclusion :

“They are … tremors of the forger when he tried to slowly simulate the signature of Wang according to his two models.” (emphasis added)

543.  The two models the judge referred to are DPH668 and 669 : Here, as can be seen, the judge was hanging his conclusion on the theory that the forger had used those two specific 1958 specimens as models for the forgery : Thereby ignoring evidence to the effect that some of the internal variations in the four suspect signatures are only found in the later samples : For example, the shape of the shelter (冖) of the radical “gwan” (軍) in “Huei” (輝) in Document A is found in a 1983 sample.  This piece of evidence from Professor Jia was never challenged.  Nor did the judge ask himself the basic question : If the forgery was done sometime between 10 April 1990 (the kidnapping) and 16 January 1998 (when the envelope was lodged in Court) as is alleged, why would the forger have chosen models more than 30 years old for his task?

544.  In my judgment Waung J was right to conclude that the judge’s finding of forgery in relation to the Wang signatures should be overturned.  He was, unfortunately, not supported in this conclusion by the majority.

Treatment of Handwriting by Court of Appeal

545.  Yeung JA said that “some of the judge’s approaches and reasoning were questionable, as was his extensive reproduction of counsel’s submissions”.  And yet he adopted essentially the judge’s approach.

546.  Yeung JA downplayed the significance of the calligraphic pen by pointing out (correctly) that there was no mention of it in the joint report.  That apart, he was in my respectful view wrong to place no weight on that aspect of Professor Jia’s evidence; and was likewise in error when he said that “by necessary implication” the judge must be taken to have rejected the argument “that the effect of a calligraphic pen explained the significant differences between Wang’s questioned signatures and his post-1967 ones”.  Professor Jia’s opinion on the importance of the broad-nib pen and the difference in use between that and a ball-point pen was never challenged by the father’s experts.

547.  As regards the signs of retouching supposedly found by the judge himself – nine in all, which no expert had found – this is what Yeung JA said :

“The further and other significant differences identified by the judge were based on common sense and/or reasonable inferences … I fail to see any merit in the suggestion that the judge ‘played expert’ as a ground to vitiate any of his findings.”

548.  There are, in my judgment, two fundamental reasons why the judge should not have “played expert” :

(1) The judge is not versed in the science of the document examiner.  In relation to a question, for example, as to whether the downward stroke of the character Wang (王) was made before or after the horizontal stroke, this is a matter of science, not of a simple observation.  The experience and expertise required for a proper evaluation is that of a document examiner, not that of a judge.  The judge had no expertise on matters such as ink infusion and the behaviour of broad-nib pens.

(2) Even when the scientific basis for an evaluation has been established, a judge is ill-equipped to make a proper evaluation.  He would not have had experience in examining suspect documents under microscopes or be aware of possible pitfalls when seeking to draw conclusions from photographic enlargements.

Yeung JA upholds the judge “playing expert” by saying that he was “entitled and justified to form his own view based on what he saw and heard in court” : If what the judge saw in court was so obvious, why was none of the nine instances of retouching put to the witnesses?

549.  Yeung JA, in his judgment, identified a number of “blatant features” in the suspect Wang signatures which were “obvious to the naked eye” : These were different from the corresponding strokes in most of the post-1967 signatures.  He said :

“I have not referred to any of the reports by the experts (at the risk of being accused of ‘playing expert’).  I do not see the need, as those differences are obvious even to untrained eyes.”

Within the narrow perspective of this statement, Yeung JA is correct.  He went on to say :

“In my view, such differences are not explainable – not by the theory of two styles of signatures, not by the suggested effect of a calligraphic pen, nor by the suggestion that Wang deliberately tried to sign in a particular classical style.

The only reasonable and logical explanation is that Wang’s questioned signatures were indeed forged signatures and modelled on his 1958 signatures.”

This is where I part company with the learned Justice of Appeal.  It is, in my judgment, not permissible to jump to a conclusion of forgery identifying points of dissimilarity without weighing all the relevant factors.

550.  The unfavourable impression Yeung JA formed of the joint report appears to have coloured his entire view of the handwriting evidence.  He said : “The blinkered approach in the joint report is surprising”.  He quoted a passage in a leading textbook as follows :

“Every questioned writing should be carefully examined with a view of determining whether by itself and without comparison with any writing it shows evidence of forgery as indicated by line quality, suspicious retouching, unnatural hesitation, pen lifts, interrupted and unnatural movements, identity of forms of several disputed signatures, or any evidence of hesitating, drawn writing.” (Questioned Documents 2^nd Edition by Albert S. Osborn- Sweet & Maxwell)”

551.  Then the learned Justice of Appeal went on to say :

“The joint report appears to have altogether overlooked the important matters identified in the quotation.”

552.  I have difficulty in understanding this criticism (which goes to the heart of the appraisal process).  Page ten (in translation) of the report says :

“The structures of the signature ‘Wang Teh Huei” appearing on the four sets of Appraisal Materials are normal, the stroke movements are natural, the degree and the pace of the pen pressure are clearly rhythmic, and the paper was even scratched and cut through when there was heavy pressure.  By making use of the stereo microscope and the infrared vidicon comparer at the Hong Kong Government laboratory for the detailed examination of the signatures ‘Wang Teh Huei’ on the four sets of Appraisal Materials, no features of imitation have been found in the writing such as pauses in or rewriting of strokes, or rigid and awkward, or apparently fast but in fact slow stroke movements.  On the whole, the level of calligraphy of the signatures ‘Wang Teh Huei’ on Appraisal Material A, Appraisal Material B, Appraisal Material C and Appraisal Material D are similar, the locations and the style features are the same; when looking at the writing feature of each character, their structure and pen movement all conform with each other and this reflects the writing habit of the one and the same person.”

553.  Another factor which greatly coloured Yeung JA’s view of Professor Jia’s evidence was the absence of any mention of matching variations in the joint report.  That was undoubtedly the case and, to that extent, does cast a shadow over the cogency of Professor Jia’s finding of authenticity.  Yeung JA said that the “theory” only gained momentum after a break in the trial for 40 days when Mr Tsui’s cross-examination was resumed; it “reached a peak” when Professor Jia gave evidence in court.  But that said, the evidence on matching variations cannot simply be dismissed as irrelevant, as the learned Justice of Appeal appears to have done : If accurate – and there was simply no challenge to that aspect of the case, as no rebutting evidence was called – then it constituted strong evidence of genuineness.  To fail to weigh it in the evidential scale was an error of law.

554.  In the end, the way Yeung JA viewed the case seems to be this : Mr Wang’s Chinese signature must have “evolved” since 1958 (as seen in DPH668 and 669); the post-1967 signatures were more “cursively, and possible speedily, written.  The stroke movements … more spontaneous; individual strokes … less clearly defined” : Hence, the supposed 1990 signatures which so closely resembled the 1958 specimens must have been forgeries by simulation.  He dismissed out of hand Professor Jia’s explanation based on different styles of signature and more deliberate strokes in using a calligraphic pen.

555.  Yuen JA, like Yeung JA, was struck by the fact that the questioned signatures closely resembled the two 1958 signatures but not the signatures since the 1980s.  She assumed that Mr Wang had discarded his old style and said there was no reason why he should revert in signing the 1990 documents.  This pre-supposes that Mr Wang had discarded his old style : As to which there simply was no evidence.

556.  Yuen JA was much impressed by the layout of the three vertically-written characters forming the Wang signatures and the way they line-up vis-a-viz a perpendicular line dropped from the top.  If one took the way the three vertical characters are structured, according to Professor Jia, the line would be drawn through the vertical stroke of the character Wang (王) : One would then find that the structure and alightment of all four sets of characters Wang Teh Huei (王德輝) conform with all the samples where the signature is written vertically.  But if one took the chart referred to in Yuen JA’s judgment, there is dissimilarity : The chart was never produced in evidence by any of the father’s experts.  It was put to Professor Jia in cross-examination : He said the lines drawn in the chart – by reference to the extremities of the three characters – were not appropriate : They did not truly represent the structure of the three characters.  There was no evidence to rebut Professor Jia’s conclusion.  The evidence concerning the alightment and structure of the characters tended to show genuineness, not the reverse : Yuen JA erred in law in placing reliance on the unproved chart.

557.  As regards the matching variations, Yuen JA said they were not of such weight as to counteract the effect of the unexplained significant differences : This was because “none of these variations in the known signatures was shown to be unique to the Deceased and discoverable only by scientific methods or experts”.  In so concluding, the learned Justice of Appeal missed much of the significance of the point : Which lies not only in the uniqueness of each variation but in the combination of characteristics.  As Waung J rightly remarked :

“… in respect of matching variations, the uniqueness of the individual feature is not the key factor but the uniqueness of the combination of features (even common) is what is important”.

558.  Yuen JA went on to say : There is no reason to assume that the person who wrote the questioned signatures had access to only some of the deceased’s known signatures.  But that is only part of the story : The forger, having planted in his mind’s eye the variations observed in the specimen signatures, must then reproduce some but not all of these features in the simulated signatures.  That would seem an astonishing feat.

559.  In my judgment, the majority finding of forgery on the Wang signatures is unsound.  What then remains of the respondent’s case?

Pulling together all the strings

560.  In this litigation, the respondent attempted to secure victory by one knock-out blow : Before the end of the year 2000 he had in his hands a report from a Mr Erich J. Speckin, a man from Michigan USA then aged about 28, to this effect : Documents A and C were written in 1996 or later : If reliable, this proved conclusively that all the 1990 documents were forged.  Mr Speckin, professing to be an expert, came to his finding by subjecting samples of the ink extracted from the documents to a series of procedures based upon the ink-drying process.  At trial Mr Speckin’s results were shown to be unreliable; his procedures conformed to no known protocol or method.  Furthermore, Mr Speckin’s qualifications turned out to have been unreliable as well.

561.  Mr Speckin’s findings in fact boomeranged on the father : It was established on Mr Speckin’s evidence that the ink on Documents A, B and C was commercially available in 1990.

562.  What then remained of the respondent’s case?

563.  The pleaded issues were a few lines of print.  Had the judge exercised proper control of the proceedings the evidence would have been confined to a narrow compass and, apart from the expert evidence on handwriting, would have occupied no more than a few days in court.

564.  The aim of any piece of civil litigation is the just, expeditious and economical disposal of the matter at hand.  The trial here failed on all three grounds.  The trial judge’s error of approach was not corrected by the majority in the Court of Appeal.  Hence it fell to this Court to view the primary facts afresh and to draw the right inferences from them.

565.  In my judgment the circumstantial evidence and the handwriting evidence, taken together, come nowhere near tipping the scales in the respondent’s favour.  The direct evidence showed Mr Wang signing the will on 12 March 1990 and entrusting custody of it to his wife sometime that month; it remained in her custody until it was lodged in court in January 1998.

566.  This appeal must be allowed and the 1990 will admitted to probate.

Sir Noel Power NPJ:

567.  I have had the advantage of reading in draft the judgments of Mr Justice Chan PJ, Mr Justice Ribeiro PJ, Mr Justice Litton NPJ and Lord Scott NPJ.  I agree with them that the appeal must be allowed, the orders of the courts below set aside and the will of 12 March 1990 admitted to probate.

568.  I wish only to state my reasons for finding that the evidence relied upon to establish forgery, particularly that of the handwriting experts, was not of sufficient cogency to establish that the deceased’s signatures were forged.  On the morning of the 7^th day of the hearing of this appeal, Mr Thomas QC for Mr Wang Snr accepted that the standard of proof referred to In re H (minors) [1996] AC 563 had application to the case which he was seeking to establish.  He said “I have accepted and I do not wish it to be thought otherwise, that I bear the burden of Re H by which I mean the standard of Re H, if I do not, I am finished.  It could not be otherwise.” (Day 7, p 54, l 21)

569.  I turn, in the light of that concession, to examine the evidence of the experts for Mr Wang Snr and those for Mrs Nina Wang.  For Mr Wang Snr they were

(i) Gus Lesnevich, an American handwriting expert.  He was in the witness box for 7 days.

(ii) David Tsui, a Chinese handwriting expert based in Canada.  He was in the witness box for 19 days; and

(iii) Mr Patrick Cheng, an independent expert from the Hong Kong Government Laboratory.  He was in the witness box for 10 days.

570.  Mr Gus Lesnevich acknowledged that he could not read, write or understand Chinese and that he did not have long professional exposure to Chinese handwriting or its comparison analysis.  It was his opinion that the 1958 signatures of Mr Wang were more controlled and more artistically written than the post-1967 signatures in which, in his view, significant deteriorations and degenerations were visible.  He was satisfied that the questioned signatures bore similarities to the 1958 signatures but were completely different from his post-1967 ones.

571.  In his opinion, the author of the questioned signatures could not have been Wang.

572.  He was also satisfied that the signatures of Mr Tse were forged.

573.  Mr Lesnevich’s report was comprehensive and fully reasoned, however, I am satisfied that caution must be exercised when assessing the evidence of Mr Lesnevich given his unfamiliarity with Chinese handwriting.  Waung J, however, puts the matter too highly when he says that he is satisfied that occidental methodology “is of doubtful value if not wholly inappropriate”.

574.  Mr David Tsui in his report dated 15 August 2000 came to the conclusion that: “It is likely that the four questioned ‘Wang Teh-huei’ signatures on exhibits (1A)-(1D) were not written by Mr Wang Teh-huei whose known signatures are contained in exhibits (2), (3), (5), (6), (9), (10) and (16).”

575.  As to the questioned signatures of Mr Tse, he said “In my opinion, the four questioned ... signatures on exhibits (1A)-(1D) were not written by Mr Tse Ping Yim”.

576.  Mr Patrick Cheng was more cautious in his conclusions.  In his first report made on 29 May 2000, he stated: “I have come to the qualified opinion that the questioned signatures pertaining to Mr WANG Teh-huei might not have been written by Mr WANG Teh-huei”.  As regards the signatures of Mr Tse, however, he stated that he was of the opinion “that the questioned signatures pertaining to Mr Tse on the four sheets of will ... were not written by Mr Tse Ping-yim”.

577.  In a later report of 6 September 2001 he stated: “In view of the enhanced evidence as a result of the additional control signatures and the concomitant reinforcement of the significance of the features, and the significance of the differences found between the questioned and control signatures, I am of the opinion that the questioned signatures were probably not written by Mr WANG.” (Emphasis supplied.)

578.  Mrs Nina Wang commissioned a report from three eminent Chinese writing experts.  It appears that all three were in court but, in the outcome, only one, Professor Jia, gave evidence.  His evidence in chief lasted 17 days and he was cross-examined for 17.5 days.  He was satisfied that the questioned signatures were genuine.

579.  At trial, Yam J accepted the evidence of the three experts called by Mr Wang Snr and rejected that of Professor Jia and the opinion in the Joint Report.

580.  This a trial judge would have been entitled to do.  Having heard evidence over many days and it would be for him to determine the weight to be given to the evidence of the experts and to apply his mind to an assessment of the handwriting evidence in the light of the evidence which he accepted.

581.  Yam J, however, in my judgment, did not follow that course.  His judgment is full of unacknowledged and unexplained copying from the submission of Mr Edward Chan SC, appearing for Mr Wang Snr.  Argumentative sentence after argumentative sentence, which would be quite proper in the mouth of counsel, were, page after page, adopted by the trial judge.

582.  The timing of the written submissions of counsel at trial is of importance in this regard.  The following chronologies are taken from the judgment of Waung J.

Chronology of Written Submissions on Handwriting

Chronology of Written Submissions on Forgery by Suspicious Circumstances and Law

583.  No one has questioned the assessment of Waung J “that in respect of the key analysis and evaluation section of the Judgment on handwriting, the copying was something like 95% or more”.

584.  Further, the extensive copying was done only from the submissions of Mr Edward Chan SC dated 20 September and 4 October 2002.  There was no copying from later submissions.

585.  The concern must be, in the light of the above, whether the handwriting evidence was ever evaluated by the trial judge.

586.  Given the unacknowledged and unexplained copying of the submissions of counsel I am left in doubt as whether the trial judge did give consideration to and make assessment of that evidence.  If he was adopting the submissions as his own he should have made it plain not only that he was doing so but also that he did so only after a careful evaluation and acceptance of them.

587.  Yeung JA dealt with the copying by the trial judge of counsel’s submissions as follows:

“366.    The conspicuous and extensive reproduction of counsel’s submissions in the judgment is unconventional, surprising, and unnecessary. If the judge accepted a party’s case entirely, he could have paraphrased its submissions and stated his agreement, and vice versa. He would not then be the subject of any criticism.

 367.    On principle, however, a judge is entitled to accept or reject a party’s case in its entirety. A judge’s total adoption or rejection of counsel’s submissions per se does not imply the lack of independent adjudication, nor does it constitute a valid ground for upsetting the judgment on the basis of an unfair trial.

 368.    The record of the proceedings clearly indicates that the judge was fully aware of and keenly interested in the issues and the parties’ arguments (as Mr. Thomas agreed).

 369.    The judge dealt with the respective cases of the parties thoroughly. He accepted most of the arguments advanced on the Father’s behalf, but he also rejected some of them.

 370.    The judge accepted the Wife’s case on the ink dating. He also dealt sufficiently with the circumstantial evidence put forward by the Wife.

 371.    That the judge did not deal explicitly with every piece of circumstantial evidence supportive of the Wife’s case does not mean that the judge was neglectful of such evidence. It was neither realistic nor reasonable to expect the judge to deal with all the points made Mr. Lee when the judge considered many of them ‘a complete waste of time’.

 372.    Moreover, the judge expressly said in his judgment (Para. 32.2), ‘However if any point which has not been mentioned herein but at the same time not in line with any of the findings or decisions I made herein, it should be taken that the point has impliedly been rejected by the Court. In fact vice versa is also true.’

 373.    The judge extensively dealt with the handwriting evidence. He clearly considered and analysed the experts’ evidence fully and carefully before he reached his conclusions on the matter.

 374.    The suggestion that the judge had abdicated his responsibility by not properly evaluating the parties’ cases and evidences (indeed a most serious allegation) is not justified.”

588.  Yuen JA dealt with this matter as follows:

“899.    Before I deal with the evidence however, I have to deal with one of the grounds of appeal raised by the Wife.  It is to the effect that the judge had failed to bring an independent mind to bear on the case and it is said that this is shown by the fact that he had copied substantial parts of the submissions of counsel for the Father.

 900.    It is clear to anyone comparing the judgment and the submissions that the trial judge did borrow substantially from them when writing the judgment.  However it does not necessarily follow from that that the trial judge had failed to bring an independent, fair and unbiased mind to bear on the case.

 901.    It would appear from the way the action was run by both sides that the trial judge had a difficult trial on his hands.  The pleadings were sketchy, both sides played their cards close to their chest, there was an abundance of evidence which came in bits and pieces  and under different conditions, and concessions which were retracted.  The evidence on virtually every point, as well as every non-point, was pursued to extremes by both sides.

 902.    After a trial which overran substantially, it was probably in the hope of producing a judgment speedily that the trial judge extracted parts of the submissions for his judgment.  I have to say that this is unusual and undesirable, but I am not persuaded that it meant that no independent thought had been exercised.  It is clear from the transcript that the trial judge followed the case closely, he probed the Father’s witnesses as hard as the Wife’s, he rejected the Father’s case on ink-dating, he refused the Father’s application to amend to plead want of knowledge and approval and he made points of his own in the judgment that were not in the submissions.  I do not therefore consider that this ground of appeal has been made out.”

589.  I cannot agree with either judge.  The trial judge failed, in my view, to demonstrate that he had made any proper evaluation of the expert evidence on handwriting.

590.  It need hardly be said that it was essential in this trial that the handwriting evidence be given proper consideration.  The trial judge failed so to do.  This failure was not acknowledged by the Court of Appeal.  The question now is whether the Court of Appeal, despite the failing of the trial judge, did, itself, made an evaluation of that evidence.  If it failed to do so, the matter comes to us without any judicial evaluation of it having been made.

591.  Yeung JA, when he came to consider the “Handwriting Evidence” began by pointing out that “both parties again presented the court with very lengthy submissions that contained many charts and diagrams and excerpts from leading textbooks on document examination.  In no less than 15 days, Mr Chang and Mr Chan both took us through their respective arguments painstakingly to demonstrate their validity.” 

592.  He went on to make a detailed examination of the submissions and expert evidence.

593.  He was satisfied that the Joint Report failed “miserably once subjected to logical analysis and careful scrutiny” and that “all in all, Professor Jia’s reasons for his opinion do not stand up to logical analysis”.  He was satisfied: “In sum, there is no convincing explanation for the ostensible similarities between Wang’s questioned 1990 signatures and his 1958 signatures despite the ‘evolution’ and for the obvious differences between Wang’s questioned signatures and his post 1967-signatures”.

594.  Yuen JA also made her own careful examinations of the evidence in the light of the evidence of the experts.  When so doing, she made no reference to the findings of the trial judge and was, in the outcome, satisfied that “the evidence shows that the signatures purporting to be those of the Deceased in the documents being propounded have been forged”.

595.  She went on to find that there was not “sufficient cogent evidence to prove that he (Tse) did not make the signatures appearing on the documents being propounded”.  She did not, however, go on to deal with the implications of this finding vis-à-vis the expert evidence generally.

596.  Waung J made his own meticulous assessment of the handwriting evidence and the experts’ opinions.

597.  There can be no doubt that by the time the appeal reached this Court the handwriting evidence had been carefully scrutinized by the Court of Appeal.

598.  One judge (Yeung JA) was satisfied that Tse’s signature was a forgery, one (Yuen JA) was not satisfied that that had been proved and one (Waung J) was satisfied that it was not a forgery.

599.  The three experts for Mr Wang Snr had all been positively satisfied that Tse’s signatures were forged.  They had, except for Mr Lesnevich, been less positive about the signatures of Mr Wang.

600.  Indeed, in the following passage from the transcript of the hearing before the Court of Final Appeal Mr Thomas QC, despite the evidence of his experts, abandoned the assertion that Mr Tse’s signature was forged:

“LORD SCOTT: Well, but you either accept that the Tse signature on the documents was his or you do not.  And you do not accept it was his signature, so therefore you are saying it was a forgery?

Mr Thomas:  may I take a moment off on that.  I took that break because I wanted to test Mr Chan’s views on the expert evidence.  And what he tells me, the answer that I give you is we do not assert that it was a forged signature.

LORD SCOTT: Thank you.

Mr Justice Chan: So you would accept the majority of the Court of Appeal’s finding that Mr Tse’s signatures on the 1990 documents were not forged?

Mr Thomas: No.  Not quite.  We no longer put forward the proposition that a person other than Mr Tse put his signature where Mr Tse’s signature appears.  Which is not quite the same thing.”  (Day 7, p 45, l 16 et seq)

601.  As can be seen from the foregoing at the conclusion of the hearing before the Court of Appeal the expert evidence called by Mr Wang Snr as to the signatures of Mr Tse was in disarray.  One judge rejected it entirely, one was not satisfied that the positive opinions of all three experts as to Mr Tse’s signatures could be accepted, only one was prepared to accept it, and, before this Court, counsel no longer contended that they were forgeries.

602.  The force of the evidence of the three experts called by Mr Wang Snr that the questioned signatures of the deceased on the 1990 documents were forgeries is now, given the above, somewhat diminished.  It has to be scrutinized with particular care and weighed against the evidence of Professor Jia, Mrs Wang’s witness.

603.  This has been done by Mr Justice Chan PJ and I will say no more than that I am in full agreement with his finding that the handwriting evidence was inconclusive and equivocal.

604.  I turn finally to what, in the outcome, became the determining evidence in this matter, i.e. the evidence in the statutory declarations of Mr Tse, the attesting witness, and, to a less extent, the evidence in the two affirmations of Mrs Wang.

605.  This evidence and its effect has been comprehensively dealt with in the judgments of Mr Justice Ribeiro PJ, Mr Justice Litton NPJ and Lord Scott NPJ, and I fully concur with their finding that the evidence of Mr Tse and Mrs Wang is conclusive in this matter.

606.  I would allow this appeal and admit the 1990 Will to probate.

Lord Scott of Foscote NPJ:

Introduction

607.  I agree with Mr Justice Chan PJ, Mr Justice Ribeiro PJ, Mr Justice Litton NPJ and Sir Noel Power NPJ that for the reasons they give, with all of which I agree, this appeal must be allowed, the orders of the courts below set aside and the will of Mr Wang Teh Huei (“Mr Wang”) dated 12 March 1990 (identified in this litigation as “document A”) admitted to probate as his last will.  I want, however, to express in my own words my reasons for coming to this conclusion.  For that purpose, I gratefully adopt and need not repeat the detailed recital of the facts and the history of this litigation that has been given by Mr Justice Ribeiro PJ. 

608.  Much of the lengthy judgments in the courts below consists of an analysis of the evidence given at trial by the respective handwriting experts, opining whether the signatures of Mr Wang on the 12 March 1990 testamentary document and on the other three documents that accompanied it were his true signatures or had been forged and whether the signatures of Mr Tse Ping Yim (“Mr Tse”), who appeared to be an attesting witness, on the testamentary document and on the other three documents were the true signatures of Mr Tse or had been forged.  The trial judge, Yam J, held that both Mr Wang’s signatures and Mr Tse’s signatures had been forged.  In the Court of Appeal a majority, Yuen JA and Waung J, overturned the finding that Mr Tse’s signatures had been forged and that reversal was accepted in this Court.  But a different majority, Yeung JA and Yuen JA, upheld Yam J’s conclusion that the Wang signatures had been forged.  Accordingly, submissions on the handwriting evidence and whether that conclusion of forgery could stand were addressed to this Court by both sides.  The submissions and the arguments have been dealt with in some detail in the judgments of Mr Justice Chan PJ and Mr Justice Litton NPJ.  I do not dissent from any of their conclusions and wish to say no more about this part of the case than that in my view the opinion evidence of the handwriting experts, taken as a whole, was equivocal, justifying neither a clear conclusion that the Wang signatures were forged nor a conclusion that could confirm the signatures as genuine.  Fortunately, however, there is other evidence that, in my view, is determinative.  But I would not wish to leave the handwriting evidence without expressing my wholehearted agreement that the manner in which Yam J dealt with Professor Jia and his evidence was unfair and unjustified. 

The issue

609.  There is in this case a single issue.  The issue is whether Mr Wang signed the document (“document A”) whereby he purported to leave the whole of his estate to his wife, Mrs Nina Wang (“Mrs Wang”).  Document A is put forward by Mrs Wang as her husband’s last will.  Probate of it is sought.  The grant of probate is, however, resisted by Mr Wang’s father (“Mr Wang senior”) who is Mr Wang’s sole beneficiary under an earlier will made in 1968.  Mr Wang senior asserts that the Wang signatures on document A and the other three documents, purporting to be the signatures of Mr Wang, are not the signatures of Mr Wang.  He asserts, in short, and has pleaded that his son’s signatures have been forged.  He asks for probate of the 1968 will.  It is now common ground that, if the signature on document A is not that of Mr Wang, the 1968 will is his last will and must be admitted to probate.  Mrs Wang put Mr Wang senior to formal proof of the 1968 will, as she was entitled to do, but there was no cross-examination of the attesting witness who supported due execution of the will by Mr Wang and, accordingly, there is no issue now about the validity of the 1968 will.  So the single issue is whether document A bears Mr Wang’s signature. 

610.  But in contested litigation single issues tend to spawn sub-issues.  And the sub-issues spawned in this litigation have been legion, some plainly relevant, some of peripheral relevance and some of no relevance whatever.  It is a legitimate criticism of the conduct of the trial that the judge failed to prevent much time being spent in the pursuit of sub-issues falling squarely within the category last mentioned.  I shall identify some of these later in this judgment. 

611.  Mr Wang was Chinese and document A was written in Chinese.  The purported date, 12 March 1990, was written in Chinese.  The purported signature of Mr Wang, Wang Teh Huei, was written in Chinese.  The validity of document A as a will did not, therefore, require any attestation of the testator’s signature.  Section 5(1) of the Wills Ordinance, Cap. 30, requires under para. (c) that, among other formalities, a testator’s signature of his will must be attested by at least two witnesses present at the time the testator signed.  But sub-section (2) says that:

“Any will of a Chinese testator written wholly or substantially in Chinese and signed by the testator shall be valid and duly executed although not executed in accordance with the rules set out in sub-section (1).” 

I have cited s.5(2) in the form in which the sub-section stood in 1990.  The section was amended in 1995.  Sub-section (1) was re-written but to much the same effect as before.  Nothing in the change of wording of sub-section (1) is material for present purposes.  Sub-section (2), however, was replaced by the following:

“A document purporting to embody the testamentary intentions of a deceased person shall, notwithstanding that it has not been executed in accordance with the requirements under subsection (1), be deemed to be duly executed if, upon application, the court is satisfied that there can be no reasonable doubt that the document embodies the testamentary intentions of the deceased person.”

The new sub-section (2) extended from Chinese wills to all wills the possibility of a will not complying with the sub-section (1) formalities being admitted to probate.  It is less clear whether the new sub-section (2) has altered the criteria with which a Chinese will must comply in order to qualify for admission to probate under the pre 1995 sub-section (2).

612.  It is not known when Mr Wang died.  He was kidnapped on 10 April 1990 and never seen again by anyone other than the kidnappers.  On 22 September 1999 Yam J granted leave to Mr Wang senior to swear to the death of Mr Wang “on or since 10 April 1990”.  So Mr Wang may have died after the new sub-section (2) had replaced the sub-section in force in March 1990.  The question whether the validity of document A as a will is to be decided by reference to s.5(2) in its 1990 form or in its amended form is, however, answered by s.30(1) of the amended Ordinance which says that the amendments:

“... shall apply to a will of a testator who dies after [the commencement of the amending Ordinance], but so that a will [made before that commencement], which, but for the provisions of this Ordinance, would be valid shall not thereby be invalidated.”

613.  The position, therefore, is that if document A can stand as a valid will under s.5(2) of the Ordinance in its 1990 form, its ability to meet the requirements of s.5(2) in its amended form does not matter, whether Mr Wang died before or after 1995.  Nonetheless, it seems to me that the implications of the 1995 amendments to s.5 have a bearing on the issue for decision in this case.  There is an equivalence between the requirement of the amended sub-section (2), namely, that “there can be no reasonable doubt that the document embodies the testamentary intentions of the deceased person” and the alternative requirements of sub-section (1).  The sub-section (1) requirements include the requirement that there are two attesting witnesses in whose presence the testator signs or acknowledges his signature and, also, the requirement that “the testator intended by his signature to give effect to the will”.  If the document signed by the testator and the attesting witnesses is testamentary in character this second requirement would normally be assumed. 

614.  The stringency of the amended sub-section (2) requirement, “... no reasonable doubt ...” etc, underlines, in my opinion, the importance both of attesting witnesses and, also, of the existence of evidence from which the inference can be drawn that the testator intended the document in question to be his will.  Sub-section (2) in its pre 1995 form does not spell out any criteria by which the ability of a document written in Chinese and signed by a Chinese person to qualify as a will should be judged.  Nor has this Court been referred to any authority on the point.  Presumably there is none.  I would, however, for my part, where probate of a document as a Chinese will is sought, look first for some independent verification that the signature on the document in question was indeed that of the alleged testator – this is the function of attesting witnesses and the evidence of a witness who saw the alleged testator sign would be the best evidence that the signature on the will was that of the testator – and, second, for evidence that justified the inference, to a fair degree of certainty, that the alleged testator regarded the document he had signed as his will.  If, for instance, the contents of the document were testamentary in character and were written by the alleged testator, this inference would usually, as in sub-section (1) cases, be clear.  But if the document had not been written by the testator, evidence might be needed that the writing of the contents had preceded the placing by the testator of his signature on the document or, in the absence of evidence of that character, some other evidence that the testator regarded the document he had signed as his will. 

615.  In the present case, the document in question, document A, is unquestionably testamentary in character.  It purports to be Mr Wang’s will dated 12 March 1990 and bears a signature which purports to be his.  But the contents of document A are not in his handwriting and the author has not been identified.  So, in my opinion, in order for the document to qualify for admission to probate as Mr Wang’s last will evidence of two sorts is necessary.  First, there must be evidence that the signature purporting to be that of Mr Wang is indeed his.  Second, there must be evidence that in signing the document, Mr Wang intended it to take effect as his will.  The evidence relied on by Mrs Wang for identification of the signature as that of her husband is, apart from the handwriting evidence of Professor Jia, the evidence of Mr Tse.  Document A, apart from the Wang Teh Huei signature, bears two other signatures.  One is a signature, written in Chinese, of the name “Tse Ping Yim”.  The other is a signature, written in English and difficult to decipher, of a name with three elements to it.  The first element is, probably, “Lee”; the second element is an initial, probably ‘K’; the third element commences with “S” and ends with, probably, “g”, but what if anything is in between is known only to the writer.  The probability is that this second signature is that of Lee Kin Sang, a Chinachem contractor, who died in 1996 before this litigation began.  So the first signatory, Mr Tse is a witness of critical importance to the outcome of this litigation. 

616.  Mr Tse had been in the employment of Mr Wang and Mrs Wang since 1983 as their butler.  On 6 September 1999, he made a statutory declaration before a solicitor, Mr Alex Lam, as to the circumstances in which he had placed his signature on document A and on the other three documents that accompanied document A.  His declaration, which he wrote in Chinese in the presence of Mr Lam, said that in March 1990, he had been asked by Mr Wang to witness his, Mr Wang’s, signature on four documents and that he had done so.  He identified his signature on each of the four documents, one of which was document A (the others were identified in the litigation as documents B, C and D).  Mr Tse’s declaration said that before he signed each of the four documents, he had seen Mr Wang sign the document.  Mr Tse repeated this evidence in a further, slightly more detailed, statutory declaration (although called an “affirmation”) that he made on 9 September 1999 before another solicitor, Mr Leung.  However Mr Tse died on 6 December 1999 and so was unable to give evidence orally at the trial.  His statutory declarations were admitted into evidence but could not be tested by cross-examination. 

617.  If the contents of Mr Tse’s statutory declarations are true, that settles the issue as to whether the signature on document A, the alleged will, purporting to be that of Mr Wang is in fact that of Mr Wang.  The pleaded allegation that Mr Wang’s signature on the alleged will (and on each of the other three documents) was forged had been made by Mr Wang senior in his Reply and Defence to Counterclaim served on 1 December 1999.  By an amendment made on 16 November 2000, after Mr Tse had died but before copies of Mr Tse’s statutory declarations confirming his signature had been supplied to Mr Wang senior, Mr Wang senior added to his pleading the allegation that the Tse signatures on the documents had been forged also.  Yam J, the trial judge, in reliance on the handwriting evidence given by the experts called by Mr Wang senior, found that that was so notwithstanding that Mr Tse, in his statutory declarations, had confirmed that the signatures were his.  In the Court of Appeal, Yeung JA upheld Yam J’s finding that the Tse signatures had been forged (para. 851 of the Court of Appeal judgment) but Yuen JA said that she was not satisfied that that was so (see paras 1020 and 1023) and Waung J, after an intensive analysis of the handwriting evidence of the respective expert witnesses, concluded that the Tse signatures were genuine and not forgeries (para. 1279).  So the Court of Appeal, by a majority, disagreed with the trial judge and held that the Tse signature on the alleged will was, as Mr Tse had confirmed in his statutory declarations, his signature. 

618.  Before this Court, counsel for Mr Wang senior, Mr Neville Thomas QC and Mr Edward Chan SC, have not sought to restore Yam J’s finding that the Tse signatures were forged.  It is now accepted that they were his genuine signatures.  But this acceptance does not of itself establish the truth of Mr Tse’s evidence in his statutory declarations that he saw Mr Wang signing the alleged will and the other documents.  It is still possible that the Wang signatures were forged and that Mr Tse added his signature to the documents at some date later than 10 April 1999 (the date of the kidnapping) in order to give support to the forgery.  But acceptance that the Tse signatures on the alleged will and the other documents are genuine shifts the focus of the inquiry.  The shift is away from the handwriting evidence and towards Mr Tse as a witness.  Mr Tse says he saw Mr Wang sign the will and that he added his, Mr Tse’s, signature at Mr Wang’s request.  If that is true, the opinion of the handwriting experts that the Wang signature is not the true signature of Mr Wang is shown to be wrong and gets Mr Wang senior nowhere.  The opinion evidence cannot possibly prevail against the evidence of, in effect, an attesting witness.  If, on the other hand, Mr Tse’s evidence is false, Mrs Wang’s attempt to propound the alleged will as her husband’s last will is exposed as a fraud and Mr Tse is implicated as a conspirator in the fraud.  The opinion evidence becomes an expensive irrelevance.  Proof of execution by Mr Wang of the alleged will depends, therefore, in my opinion, on the view taken of the evidence of Mr Tse.  Either he is a genuine attesting witness or he is a dishonest conspirator, willing to assist Mrs Wang in a conspiracy of forgery and fraud. 

619.  What of the second requirement?  If Mr Wang did sign document A (and the other three documents), does it appear that in doing so he intended to give effect to document A as a will?  If the premise that Mr Wang signed document A is accepted, the inference that in doing so he intended the document to constitute his will is overwhelming.  First, the contents of document A itself, fortified by the complementary character of the precatory instructions contained in documents B and C, establish the testamentary character of document A.  Secondly, the request by Mr Wang that Mr Tse witness his signature suggests a testamentary intention on Mr Wang’s part.  But this, of course, depends upon the contents of Mr Tse’s statutory declarations being true.  Thirdly, Mrs Wang gave evidence in affirmations that she swore in interlocutory proceedings in the probate action that preceded the trial itself that Mr Wang, on or shortly after 12 March 1990, had handed her a sealed envelope to look after for him, telling her that the envelope contained his will and confidential instructions about his estate.  Mrs Wang produced this sealed envelope to Yam J, in obedience to a court order to do so.  It was, at the direction of the judge, opened on 6 September 1999 and its contents inspected.  It was found to contain documents A, B, C and D with a Wang signature and a Tse signature on each.  If Mrs Wang’s evidence in her affirmations of the circumstances in which the envelope came into her possession is true, it confirms that Mr Wang regarded document A as his will.  The status and weight of this affirmation evidence is in issue since, in circumstances I shall have to describe, Mrs Wang was not cross-examined. 

620.  In summary, the single issue in this case turns, in my opinion, on the view taken of the evidence of Mr Tse contained in his two statutory declarations.  Second in importance is the view taken of, and the weight to be attached to, the sworn evidence of Mrs Wang, contained in her affirmations, of the circumstances in which she came into possession of the envelope containing documents A, B, C and D.  Before dealing with those determinative matters, I want to say something about the burden and standard of proof and the use of presumptions in a case such as this. 

The burden and standard of proof

621.  Mrs Wang is seeking probate of document A as her husband’s last will.  She must satisfy the court that the document was duly executed.  The concept of due execution means, in relation to a will to which English law applies, that the will has been executed by the purported testator in accordance with the prescribed statutory requirements.  These requirements, prescribed by s.9 of the Wills Act 1837, as amended (most recently by the Administration of Justice Act 1982), are identical to the requirements prescribed by sub-section (1) of s.5 of the Wills Ordinance in its post 1995 form.  For deaths before 1 January 1983 the prescribed requirements for a will under English law were the same as those prescribed by sub-section (1) of s.5 of the Wills Ordinance in its pre 1995 form.  The person seeking probate of a will has the burden of satisfying the court that the statutory requirements have been complied with.  The court must be so satisfied to the standard of certainty applicable generally to civil proceedings, that is to say, on a balance of probabilities.  All of this is trite law and the textbooks abound with illustrations produced by decided cases of circumstances where the burden of proof has been discharged and of circumstances where it has not.  But the decided cases all relate to the English law statutory requirements, or, from a Hong Kong perspective, to the s.5(1) requirements.  This case is in different territory.  Here, probate is sought of a Chinese will under s.5(2).  In such a case what does the propounder of the will have to prove?  I have already referred to this issue and repeat that, in my view, the propounder must satisfy the court that the signature on the document in question is that of the deceased and that, by signing, the deceased intended the document to take effect as his will.  Here, too, the civil standard of proof will be applicable. 

622.  So Mrs Wang had to satisfy the court that, on a balance of probabilities, the Wang signature on document A was that of her husband.  But, of course, the issue is complicated by the existence of documents B, C and D.  On each of these, too, the Wang Teh Huei signature appears.  These signatures stand or fall together.  If any is genuine, all are genuine.  If any is forged, all are forged.  It is not remotely conceivable that one may be forged and the others genuine, nor vice versa. 

623.  Mrs Wang must also, in my opinion, satisfy the court, to the same standard of proof, that the deceased, when appending his signature to document A, intended the document he was signing to be his will. 

624.  The burden and standard of proof to which I have referred would have been applicable if Mr Wang senior had simply put Mrs Wang to strict proof of due execution of the 1990 will.  But Mr Wang senior went further.  He denied that the 1990 documents had been duly executed and pleaded that the Wang signatures had been forged.  He also, in the knowledge that Mr Tse was dead and unaware that Mr Tse had before his death confirmed his signatures by statutory declaration, added to his pleading the allegation that Mr Tse’s signatures were forged.  The evidential burden of proving these positive allegations fell fairly and squarely on Mr Wang senior who had made them.  He did not plead that Mrs Wang was the forger but it was plainly his case that she had procured the forgery and was attempting to obtain probate as a will of a document she knew to have been forged.  As to Mr Tse, once it became known that he had confirmed that the Tse signatures were his signatures, the allegation of conspiracy to which Mr Tse was a party became a necessary one if Mr Wang senior’s case was to be maintained.  If the Tse signatures were forgeries, then the Tse statutory declarations purporting to confirm the signatures and stating the circumstances in which Mr Wang said he had signed the documents must have constituted perjured evidence.  If the Tse signatures were genuine then, if Mr Wang senior’s case was to be maintained, Mr Tse must from the outset have been a participator, a co-conspirator with Mrs Wang, in the scheme to obtain probate of a forged will.  In respect of all these allegations, the evidential burden of satisfying the court that they were true falls on Mr Wang senior. 

625.  Where allegations of this character are made, the courts rightly demand a standard of proof commensurate with the seriousness of the allegations.  The remarks of Lord Nicholls of Birkenhead in In re H (Minors) [1996] AC 563 at 586 are often cited and are very pertinent for present purposes.  They repay repeating:

“The balance of probability standard means that a court is satisfied an event occurred if the court considers that, on the evidence, the occurrence of the event was more likely than not.  When assessing the probabilities the court will have in mind as a factor, to whatever extent is appropriate in the particular case, that the more serious the allegation the less likely it is that the event occurred and, hence, the stronger should be the evidence before the court concludes that the allegation is established on the balance of probability ... Built into the preponderance of probability standard is a generous degree of flexibility in respect of the seriousness of the allegation.

Although the result is much the same, this does not mean that where a serious allegation is in issue the standard of proof required is higher.  It means only that the inherent probability or improbability of an event is itself a matter to be taken into account when weighing the probabilities and deciding whether, on balance, the event occurred.  The more improbable the event, the stronger must be the evidence that it did occur before, on the balance of probability, its occurrence will be established.  Ungoed-Thomas J. expressed this neatly in In re Dellow’s Will Trusts [1964] 1 W.L.R. 451, 455: ‘The more serious the allegation the more cogent is the evidence required to overcome the unlikelihood of what is alleged and thus to prove it.’”

The same thought was expressed by Sir Anthony Mason NPJ in HKSAR v. Lee Ming Tee (2003) 6 HKCFAR 336 at para. 136 when he referred to matters that “on their own raise a case of suspicion against the SFC and are capable of generating an adverse inference of ulterior intention or purpose ...” and went on:

“But when due weight is given to the seriousness of the conduct alleged against the SFC, the explanations offered on behalf of the SFC and the strong possibility that any errors of judgment on the part of the SFC were innocent, one cannot be satisfied that the ulterior intention or purpose is made out or that the inference sought to be drawn by the respondent is compelling.”

626.  The passage from Lord Nicholls’ opinion in In re H that I have cited stresses the need to concentrate on the inherent probability or improbability of the serious allegation being true.  In the present case there are two such allegations.  The principal allegation is that Mrs Wang procured the forgery of her husband’s signature on each of the four 1990 documents.  The other allegation is that Mr Tse became a conspirator with Mrs Wang in a conspiracy to utter a forged will.  The probability of these allegations being true must be judged on the evidence adduced in the case.  But it must also take account of propensity.  If such an allegation is made against a person with a record of involvement in forgery or fraud, the strength of the other evidence necessary to satisfy the balance of probability test is obviously less than would otherwise be required.  Evidence of propensity must go into the balance.  In the present case, however, there is no such evidence to go into the balance either against Mrs Wang or against Mr Tse.  Evidence to a very high standard of cogency indeed is necessary before the court can be justified in finding either to be dishonestly involved in a conspiracy to promote a forged will. 

627.  Yam J, in Chapter 5 of his judgment required Mrs Wang to “dispell (sic) suspicious circumstances” before she could succeed in obtaining a grant of probate of document A.  He identified nine categories of “suspicious circumstances” which he concluded she had failed to dispel (see paras 5.4 and 5.127 of his judgment).  This approach constituted a serious misdirection.  Mrs Wang did not have to dispel suspicious circumstances i.e. dispel the suspicions that the judge thought to be raised by the circumstances in question.  The only case pleaded against her was that of forgery and she was entitled, to use the language of the criminal courts, to the presumption of innocence.  The expression “suspicious circumstances” demonstrates, in my opinion, the judge’s error.  What was the suspicion?  Suspicion that the Wang signatures were forged is the proposition implicit in the expression.  It is true to say that some of the so-called suspicious circumstances were consistent with that proposition.  But none were inconsistent with the signatures being genuine.  The judge started from the wrong standpoint.  Instead of requiring Mrs Wang to dispel his suspicious circumstances, he should have considered all the reasonably probative circumstantial evidence in the round and asked himself whether its weight was sufficient to establish the very serious allegations that were made against Mrs Wang and Mr Tse. 

628.  Both Yeung JA and Yuen JA in the Court of Appeal made, in my respectful opinion, an associated although different error.  Both took the view that there was, so-to-speak, a split standard of proof.  Mr Wang senior had to prove to an In re H standard of probability that the Wang signatures were forged.  But even if Mr Wang senior failed to do so, Mrs Wang might nonetheless fail to discharge her burden of satisfying the court that, on a balance of probabilities, the Wang signatures were genuine.  In which case she would fail to satisfy the court that probate of the 1990 will should be granted (see Yeung JA, paras 415 and 416 and Yuen JA, para. 874).  This evidential paradox cannot, in my opinion, arise in the present case.  Mrs Wang, in seeking probate of the 1990 will, must satisfy the court that the Wang signatures are indeed his.  A widow who lived with her husband for thirty-five years odd up to the time of his death (or disappearance) could normally be expected to recognise his signature and to satisfy the court about its genuineness without the need for the opinions of handwriting experts.  In the present case, however, forgery was alleged and handwriting experts were enlisted to try to prove or disprove the allegation.  Circumstantial evidence, too, was relied on for its cumulative probative value.  But let it be supposed that this evidence does not succeed in establishing forgery.  What is left to oppose the statement by the widow that the signatures are those of her husband?  Nothing, bar the suspicions engendered by an allegation of forgery that has not been proved.  The approach of Yeung JA and Yuen JA in the Court of Appeal was, in my opinion, wrong in law.  They had both paid lip-service to the In re H requirement but, in practice, had left Mrs Wang, in order to obtain probate of the 1990 will, with the burden of dispelling suspicion of forgery derived from circumstances that did not themselves prove any such thing.  They required her, in practice, to satisfy them that Mr Wang senior’s case against her of forgery was wrong. 

629.  Waung J in the Court of Appeal gave a dissenting judgment.  He disagreed with the conclusions drawn by Yam J and by the majority in the Court of Appeal from the evidence of the handwriting experts and he regarded Yam J’s “suspicious circumstances” as “wholly irrelevant to execution and to forgery”.  He drew a distinction, with which I respectfully agree, between cases where the only issue is whether the will was properly executed and cases where the issue is whether the testator had knowledge of and had approved the contents of the will.  “Suspicious circumstances” with which the propounder of the will, or the beneficiary under the will, is associated may require the propounder of the will to dispel suspicions that the testator did not have that knowledge and had not given that approval.  But a case of lack of knowledge and approval must be pleaded.  In the present case, lack of knowledge and approval by Mr Wang of the contents of document A was not pleaded and so, it is common ground, was not an issue.  I agree with what Waung J had to say on this point and am also in respectful agreement with what Mr Justice Ribeiro PJ has said about it in his judgment on this appeal.  But Waung J then went on to discuss the omnia praesumuntur rite esse acta presumption and held, first, that the presumption was applicable to Chinese wills probate of which was being sought under s.5(2) of the Ordinance (see para. 1069) and, more particularly, that the presumption was applicable to the question whether the 1990 will had been signed by Mr Wang.  I am afraid that I can agree with neither of these propositions. 

630.  The omnia praesumuntur presumption “applies where the will is regular on the face of it, with an attestation clause and the signatures of the testator and witnesses in their proper place ...” (see Halsbury’s Laws of England, 4th Ed. Vol. 17(2) at para. 304, cited by Waung J at para. 1062 of his judgment).  But if the issue in the case is whether the signature purporting to be that of the testator is in fact that of the testator, the condition for the application of the presumption is not satisfied.  This would particularly obviously be so in the case of a Chinese will where attesting witnesses are not necessary.  If the testator’s signature is challenged the propounder of a Chinese will cannot, in my opinion, simply rely on the presumption and call on the objector to rebut it.  The propounder must produce at least some evidence that the signature is that of the testator before the will can be admitted to probate in solemn form.  Waung J suggested, in para. 1069, that by not applying the presumption the court would be frustrating the legislative intention behind s.5(2) in its pre 1995 form.  Not so.  Section 5(2) makes clear that an attesting witness is not essential; it does not avoid the need for evidence of some sort that the signature is that of the testator.  Evidence from a family member, or indeed from anyone, familiar with the deceased’s signature might, for example, suffice for s.5(2) purposes. 

631.  In my opinion, the omnia praesumuntur presumption has no application to the present case nor to any case where the signature of the testator on a Chinese will is disputed.  Evidence of probative value that the signature is that of the deceased is needed. 

632.  I can now turn to Mr Tse’s evidence and Mrs Wang’s evidence. 

Mr Tse’s evidence

633.  Why was Mr Tse’s evidence, contained in the two statutory declarations, rejected?  Why did Yam J and the majority in the Court of Appeal prefer to treat him as a fraudulent conspirator rather than as an honest witness?  Let me start with Yam J.  There are several features of Yam J’s judgment regarding Mr Tse that deserve mention.  I do not list them in any particular order of importance. 

(1) Yam J took the view that Mr Tse favoured Mrs Wang and had been favoured by her rather than by Mr Wang.  Thus: “it is quite obvious that he felt he was much better treated by Mrs Wang” (para. 5.76) (I have been able to find no evidential basis for this remark); “he got his job as a butler because of Mrs Wang” (para. 5.77); “Teddy was not particularly polite to Tse.  He called him ‘old chap’ ...” (para. 5.81) and “... given the talkative nature of Mr Tse, it was unlikely that he would be able to win the heart of Teddy.  It was thus ... more than unlikely that Teddy would ask him to be a witness to his will” (para. 5.82).  These are inconsequential trivia incapable of providing any basis for the surmise that the relationship between Mr Wang and Mr Tse was other than that to be expected between an employer and a senior domestic servant of long standing.  The judge’s proposition that the relationship made it unlikely that Mr Wang would have asked Mr Tse to witness his will is flimsy to a degree that deprives it of any useful substance.  It is, as the cases show, very common for a domestic servant to be asked to witness a will signed by the testator or testatrix in his or her home.  Indeed a domestic servant is often the obvious person to be the witness since members of the family are likely to be beneficiaries. 

(2) In the first of his two statutory declarations, made on 6 September 1999 in the Chinachem office premises in the presence of Mr Alex Lam, the solicitor who had gone to the premises to take the declaration, Mr Tse referred to the riding accident that Mr Wang had suffered two days earlier on 10 March, to Mr Wang’s two days in hospital and to the loving relationship between Mr and Mrs Wang.  The judge’s comment was:

“It is quite obvious that he must have been told to put all these information in to assist his lady boss to enable her to run a case that the riding accident triggered off Mr Wang to make a new will.  Who would have told him that?  His reference to the loving relationship between Mr and Mrs Wang was obviously the result of some coaching so as to assist Mrs Wang’s case to show some reason for Mr Wang’s intention or motivation to give all his estate to her.” 

This comment strikes me as unfair to Mr Tse, unbalanced and unjudicial.  Mr Tse had been asked by Mr Lam to describe “how these four questioned documents came about”.  The riding accident on 10 March which led to Mr Wang being concussed and spending two days in hospital must have been a relatively traumatic event not only for Mr Wang himself but also for the members of his household.  Mr Tse had spent a number of hours sitting at Mr Wang’s bedside in the hospital.  Mr Wang discharged himself from the hospital, returned to his office and on the same day the signing of the documents took place.  It seems to me entirely natural that Mr Tse should have associated the events of 12 March with the riding accident on 10 March and the two days in hospital.  The judge’s comment suggests very strongly to me that he had already decided that Mrs Wang was putting forward a fraudulent case and was intent on reconciling the Tse evidence with that premature premise.  As to the reference made by Mr Tse to the relationship between his employers, there had been media comments about the impending probate litigation and about allegations by Mr Wang senior that Mrs Wang was going to put forward a fraudulent case in order to obtain her husband’s estate.  Against this background of public speculation, Mr Tse’s references to Mr and Mrs Wang’s relationship does not seem to me to warrant the suspicions the judge derived from them. 

(3) Yam J concluded that Mr Tse’s 6 September statutory declaration had been “already prepared or written by him before Lam’s arrival at Chinachem’s office ...” (para. 5.88).  This conclusion is contradicted by Mr Lam’s oral evidence that he had seen Mr Tse writing out the declaration, that it had taken an hour or so for Mr Tse to do so and that neither he, Mr Lam, nor anyone else had spoken to Mr Tse while he was writing the declaration.  Yam J held that Mr Lam was not a credible witness (para. 5.97) but Yeung JA described his criticism as harsh (para. 482), Yuen JA said that “where there was no real challenge to [Mr Lam’s evidence], it should not have been rejected on the basis of unreliability” (para. 961) and Waung J expressed the same view (para. 1087).  There was no evidence which justified Yam J in drawing the inference that Mr Tse’s 6 September declaration had been pre-prepared. 

(4) In para. 5.90 of his judgment Yam J said that it was “clear that since the disappearance of Mr Wang, Mrs Wang had been treating Tse extremely well”.  The plainly intended innuendo was that Mrs Wang had been rewarding Mr Tse for his participation in her endeavour to produce a forged will.  The evidential basis for this was that Mr Tse’s wife became in 1992 one of two shareholders of Global Cheers Ltd, a company holding contracts to perform cleaning services in buildings developed by Chinachem.  It appears, however, that Mrs Tse was merely a nominee and that in a number of companies in the Chinachem group the shares were held by nominees for the beneficial owner or owners.  Yam J’s comment about this practice, couched in terms intended to fortify a suspicion, lacking any evidential foundation, is to be found in para. 5.92 of his judgment and is not worth repeating.  The comment is worthless.  It must be remembered, moreover, that in September 1999 Mr Tse was nearly 70 years old and had been for many years a trusted senior domestic servant of Mr and Mrs Wang and, after April 1990, of Mrs Wang alone.  That Mrs Wang should have treated him with some indulgence is not a reasonable ground for suspicion.  It is only if forgery on her part, assisted by Mr Tse, is assumed that the indulgence can be categorised as suspicious.

(5) The final matter relied on by Yam J for justifying his rejection of the evidence contained in Mr Tse’s statutory declarations was that on 24 September 1999, the day after Mr Wang senior had been granted leave by Yam J to report to the police his allegation that the Wang signatures on the four 1990 documents had been forged, Mr Tse left Hong Kong and went to mainland China.  This was represented by the judge, and by counsel for Mr Wang senior, as Mr Tse fleeing the jurisdiction in order to make himself unavailable for questioning about the forgery.  This inference, and the innuendo underlying it of complicity by Mr Tse in the alleged forgery, lacks cogency for a number of reasons.  First, let me rehearse the relevant dates.  On 14 September 1999 Mr Tse left Hong Kong for mainland China.  He returned to Hong Kong on 18 September but went back to the Mainland on 24 September.  He was building a house for himself in a village about 250 kilometres from Shenzhen.  On 14 November he went to the outpatients department at Shenzhen Peoples Hospital.  He was kept in the hospital and, after a CT scan, cancer of the liver was diagnosed.  He stayed in the Shenzhen hospital until 4 December when he returned to Hong Kong and was admitted to the Princess Margaret Hospital where he died on 6 December 1999.  It is important to note that there is no evidence that he, or anyone else, realised that he was suffering from a terminal cancer until the diagnosis made at the Shenzhen Peoples Hospital.

Mr Tse was Mrs Wang’s most valuable witness.  He was expected and expecting to give oral evidence at the trial.  His death before the trial must have been a surprise and a blow for Mrs Wang, and, correspondingly, a surprise and a boon for Mr Wang senior.  Over the period between 6 October 1999, when Mr Wang senior made his forgery complaint to the police and 4 December 1999 when Mr Tse entered the Hong Kong hospital and, apparently, the police went to the hospital to see if they could interview him, there is no evidence of any attempt by the police to contact him.  They could easily have found out his address and written to him to arrange an interview.  In the period after 6 September 1999, when Mr Wang senior and his solicitors had become aware that Mr Tse’s signature was on the four 1990 documents, there is no evidence of any attempt by Mr Wang senior’s solicitors to question Mr Tse about the 1990 documents and his knowledge of the signatures on them.  The suggestion that Mr Tse absented himself from Hong Kong, or that his absence from Hong Kong was procured by Mrs Wang, in order to avoid him being questioned cannot, in the absence of evidence of any such attempts and their failure, be made good.

634.  Yeung JA, who gave the first judgment in the Court of Appeal, said that “the circumstances in which Mr Lam took the statement from Tse on 6 September 1999 were indeed highly suspicious.  Even more so was Tse’s sudden and unexplained departure from Hong Kong” (para. 482) and concluded that there were “valid reasons for rejecting Tse’s statements” (para. 484).  The matters he relied on for reaching that conclusion were as follows:

(1) Yeung JA said that it was “inconceivable” that Mr Tse could have written out the statements without prior coaching because “the contents of [the statements] were a perfect match with the wife’s case” (para. 502).  But this completely begs the question.  If the statements were made without coaching and were true, of course the contents would represent Mrs Wang’s case.

(2) Yeung JA said that the four questions put to Mr Tse by Mr Lam could not have resulted in Mr Tse producing without coaching his 6 September 1999 statement.  I do not agree.  Mr Lam’s third question required Mr Tse to describe the circumstances which had led up to the production by Mr Wang of the four 1990 documents and Mr Wang’s request that Mr Tse witness his, Mr Wang’s, signature on them.  It does not seem to me at all unnatural that Mr Tse would regard the riding accident on 10 March as the start of the narrative.  Yeung JA commented, also, that there were “strong internal marks of untruth, unfairness and partiality in Tse’s statements” (para. 503).  He did not identify the passages he had in mind and, for my part, I am unable to detect any that attract those descriptions. 

(3) Yeung JA criticized the fact that Mr Tse, when making a statement to the police about Mr Wang’s kidnapping on 10 April, had said nothing about the signing of the documents on 12 March.  He said it was “highly improbable” that “a matter of such peculiarity and immediacy should have escaped Tse’s mind” (para. 506).  I regret to say that I do not follow the reasoning.  What is so obviously peculiar about the witnessing of Mr Wang’s signature on four documents?  Mr Tse probably noticed that document A was a will, for the document so stated immediately above the place where Mr Tse signed.  But he did not read any of the documents and, even if he had, the connection with the kidnapping that seems so apparent to the judge could well have escaped him, as it escapes me.

(4) Yeung JA referred to Mr Tse’s “highly suspicious” departure from Hong Kong on 24 September 1999 (para. 518) and drew the inference, which he described as “compelling” “that he wanted to avoid interview by the police and that those controlling him also wanted to prevent him from such interview” (para. 521).  As I have already said, there is no evidence that the police made any attempt to interview him before his return to Hong Kong on 4 December.  There is no evidence that he intended when he left Hong Kong on 24 September to remain out of Hong Kong for an extended period in order to avoid police questioning.  There is every reason to believe that he intended to return to Hong Kong to give evidence at the trial on the lines of his statutory declarations.  And there is no evidence that Mrs Wang was “controlling him” (for it must be she that Yeung JA had in mind) to prevent him from being interviewed.  No one, I repeat, ever asked to interview him or took any steps prior to 4 December to attempt to interview him.  The inferences drawn by Yeung JA from Mr Tse’s departure for the Mainland on 24 September are, in my opinion, based on speculation bred from an assumption that Mrs Wang had procured a forged will.  If the will was genuine, there is nothing suspicious in Mr Tse’s departure for the Mainland on 24 September.

635.  Yuen JA, in para. 953, expressed the view, with which I entirely agree, that Mr Tse was the most important witness in the case.  She disagreed with Yam J’s rejection of the evidence given by Mr Alex Lam about the circumstances in which Mr Tse had written out his 6 September 1999 statement.  I respectfully agree with her as to that also.  So one must approach Mr Tse’s 6 September statement on the footing that it was written out by him, over the period of an hour or so, without assistance from anyone.  Nonetheless Yuen JA concluded (paras 979 and 980) that “it would be unsafe for the court to act on [Mr Tse’s] statements alone”.  There appear to have been two main reasons, and a subsidiary reason, why she came to that conclusion. 

(1) The first of her main reasons was that Mr Tse had volunteered information about the loving relationship between Mr Wang and Mrs Wang.  Why had he volunteered this information?  It was not pertinent, she said, to the subject-matter of the statement (paras 964 and 965).  Yuen JA was, obviously, drawing the inference that Mr Tse had been coached by Mrs Wang to add some such passage to his statement.  It was certainly the case that Mr Lam had not asked Mr Tse to comment on the relationship between Mr and Mrs Wang.  And I agree that the comments he made were, therefore, volunteered by him rather than sought by Mr Lam.  But I do not agree that that carries with it the inference that he had been coached by Mrs Wang as to what to say.  He could not have been unaware why he was being asked to make the statement.  There had been media interest in the probate dispute between Mrs Wang and Mr Wang senior and whether the latter or the former was entitled to Mr Wang’s very considerable fortune.  He was a senior domestic servant of the family and it would have been extraordinary if he had not been aware of the media speculation about the dispute.  Moreover, human nature would have made it almost inconceivable that Mr Tse did not read the contents of the documents before commencing to write his statement.  If he had read the documents, as he is very likely to have done, the contents of A and D in particular might well have prompted him to think that that the relationship between Mr and Mrs Wang would be a relevant matter to mention.  And although Mr Tse’s comments about the relationship were volunteered, no one has suggested they were untrue.  Other witnesses said much the same.  The inference, therefore, that Mr Tse made these remarks because he had been coached to do so was not, to my mind, a compelling one.  I do not, therefore, regard this particular matter as casting a shadow over the credibility of the contents of Mr Tse’s statements. 

(2) The second of Yuen JA’s main reasons was Mr Tse’s departure from Hong Kong on 24 September 1999.  In para. 978 she said that by his departure he “made himself unavailable to the Hong Kong police” and that he “never allowed himself to be questioned by an independent investigator”.  I do not think either of these conclusions was justified by the evidence.  As I have already commented, there was no evidence of any attempt made either by the Hong Kong police or by Mr Wang senior’s solicitors to interview Mr Tse in the period between his departure from Hong Kong on 24 September 1999 and his return on 4 December.  If there had been some such attempt and Mr Tse had refused to make himself available, Yuen JA’s remarks might have been justified.  But not otherwise.  It has not been shown that it was his departure from Hong Kong that prevented his being questioned (see para. 979).  It was his death that did that.  The inference I would draw from the relevant facts is that at the time he left for the Mainland there was every expectation that he would return to Hong Kong to give oral evidence at the trial of the probate action.  There is no evidence that he or anyone else knew that he had a terminal cancer. 

(3) The subsidiary reason why Yuen JA was unenthusiastic about accepting as true Mr Tse’s 6 September statutory declaration seems to have been the literate and fluent style in which it was written (see para. 963): “The first point that strikes the reader is that it is very well-structured and composed” and “it remains a remarkable work for an unscripted first draft”.  I may be doing Yuen JA an injustice in reading into these comments some doubt that Mr Tse, a butler, could unassisted have produced such a work.  If I am, I must apologise and note that the literacy and fluency of the composition is no ground for suspecting Mr Tse of fraudulent conspiracy. 

636.  I have spent some time in examining the reasons why Yam J and the majority in the Court of Appeal rejected as not credible, or felt unable to accept the truth of, Mr Tse’s two statutory declarations.  None of the reasons, individually or cumulatively, seems to me to have much weight.  Put into the balance on the other side the improbability of Mr Tse committing himself to the fraudulent conspiracy in which he would have to have involved himself if his statutory declarations were untrue.  Nothing in his past history or known character apt to reduce that improbability has been suggested.  No evidence has been adduced of any unexplained additions to his wealth or to that of his family members.  In my judgment the grounds on which Mr Tse’s evidence was rejected, or not accepted, cannot withstand critical examination.  I think his sworn evidence about the circumstances in which he came to put his signature on the 1990 documents should have been accepted.  And, as I have said, if accepted that evidence is determinative of the case. 

Mrs Wang’s evidence

637.  Mrs Wang had given sworn evidence in two affirmations of the circumstances in which documents A, B, C and D had come into her possession.  In para. 9 of an affirmation she swore on 16 January 1998 she said that after his riding accident her husband had told her he had made a new will and that a few days later

“... he gave me a sealed envelope for safe-keeping and he told me that the envelope contained his will together with confidential instructions relating to his estate.” 

On the same date, 16 January 1998, she lodged the envelope, unopened, with the court.  It was opened at the direction of Yam J on 6 September 1999, nearly 20 months later.  Mrs Wang’s affirmation was sworn in the first probate action (No. 6 of 1997) commenced by Mr Wang senior on 28 May 1997 but struck out by the Court of Appeal on 14 January 1999 as misconceived.  Mr Wang senior, having obtained on 22 September 1999 the leave of the court to swear to the death of Mr Wang, commenced on 24 September 1999 the probate action (No. 8 of 1999) that is now before this court.  On 18 November 1999 Mrs Wang swore in this probate action an affirmation of testamentary scripts to which she exhibited a copy of the 16 January 1998 affirmation and in which she repeated and confirmed the truth of the contents of that affirmation. 

638.  So there was sworn testimony from Mrs Wang, on file in this probate action, dealing with the circumstances in which the envelope containing the four 1990 documents had been handed to her by her husband. 

639.  In addition a witness statement was prepared setting out the evidence proposed to be given in chief by Mrs Wang at the trial of this action.  The trial lasted 172 sitting days.  On Day 151, a discussion took place in chambers before Yam J about the future progress of the trial and, in particular, about when Mrs Wang might go into the witness-box and how long her evidence would take.  Mr Edward Chan SC, who led for Mr Wang senior at the trial, is recorded as saying that the duration of his cross-examination of her was likely to set the record for the trial (see p.113 of the transcript of Day 151).  The lengthiest cross-examination so far had been Mr Martin Lee SC’s 20 day cross-examination of one of Mr Wang senior’s witnesses, Mr Yih Lee Kong.  Yam J is then recorded as expecting a further witness statement from Mrs Wang, no doubt on the footing that her existing witness statement did not cover all the matters raised in the course of the evidence given by the witnesses already called and which the judge expected her to want to deal with.  However on Day 169 Mr Lee informed the judge that Mrs Wang was not going to be called as a witness and that he was about to close his case (p.11 of the transcript of Day 169).  A discussion about the witness statements of those persons who had not been called as witnesses then followed.  There were, I think, 11 of them, including Mrs Wang.  Yam J directed that “All witness statements not called will be expunged, unless they are agreed to”.  The next day, Day 170, Mr Lee made further submissions about Mrs Wang’s evidence, with particular reference to her two affirmations.  It was agreed that any further submissions from him and Mr Chan dealing with the matter would be put in writing and supplied to the judge (see p.7 of the transcript of Day 170).  In the event no such written submissions were produced.  There the matter rested until the judge gave his judgment. 

640.  Yam J in his judgment made at least two references to the evidence Mrs Wang had given in her affirmations about the manner in which the envelope containing the 1990 documents had come into her possession (see paras 5.35 and 5.118).  In para. 5.118 the judge cited a passage from her 16 January 1998 affirmation that included the passage I have cited (para. 634 above).  In para. 5.120 the judge cited also from two other of Mrs Wang’s affirmations.  It is plain that in so doing the judge was treating the contents of the affirmations to which he was referring as evidence before the court.  Mr Tang SC, counsel for Mr Wang senior in the Court of Appeal, conceded that the court was entitled to refer to the contents of the affirmations “as the court’s approach was inquisitorial in nature” (para. 498). 

641.  But Yam J, and Yeung JA and Yuen JA in the Court of Appeal, treated Mrs Wang’s affirmation evidence of the provenance of the envelope containing the 1990 documents as deprived of any evidential weight because, it was said, she had not allowed herself to face cross-examination.  Yam J, in para. 5.122 of his judgment, posed certain questions about the envelope and its contents and then said:

“As Nina did not come forward to give evidence, we would not know what would be her answer to these two questions.”

He then answered the questions in a sense adverse to her.

642.  Yeung JA dealt with Mrs Wang’s affirmations in paras 495 to 499 of the Court of Appeal judgment.  He said that “the affirmations were available to court only as affirmations of testamentary scripts, not as evidence for the purposes of the trial” (para. 495).  This was, in my opinion, an error.  Yam J had, in his judgment treated the contents as evidence before the court, as he had power to do (see O.38, r.2 and note 38/2/5 of the White Book) and as counsel for Mr Wang senior had conceded he could.  Yeung JA referred (in para. 499) to the value of cross-examination as a “legal engine ... invented for the discovery of truth” and went on (para. 500):

“By choosing to avoid this ‘legal engine’, the Wife had deprived the court of one of the major means to test the veracity of her assertions.  While the court was entitled to refer to her affirmations, I am of the view that little weight should or could be attached to them.” 

And Yuen JA, in paras 986 to 990, made the same point.  She said “... a person who chooses not to give evidence and be cross-examined has to bear the consequences of that decision” (para. 989) and that “... by not offering herself as a witness at trial at all ...”.  Mrs Wang was avoiding cross-examination (para. 990). 

643.  In my opinion, the remarks of each of these judges about Mrs Wang “avoiding” cross-examination were based on a fallacy.  It was not open to Mrs Wang to “avoid” cross-examination.  She had given in her affirmations evidence which had obvious probative relevance to the provenance of the will of which probate was sought.  There were two witnesses who appear on the face of the will to have witnessed the testator’s signature.  Mr Tse died before trial and so could not be cross-examined; the other, whose identity was never firmly established, but who was probably Lee Kin Sang, had died before the probate proceedings had started.  So the provenance evidence given by Mrs Wang in her affirmations was of enhanced importance.  It is well-established that probate proceedings where probate of a will in solemn form is sought are not straightforward adversarial proceedings between the parties to the litigation.  The function of the court has an inquisitorial element to it.  As Cairns J said in In re Muirhead [1971] P 263 at 265:

“... it is the duty of a Court of Probate to give effect, if it can, to the wishes of the testator as expressed in testamentary documents.”

It is accepted, and well established, that attesting witnesses are not the witnesses of any party, but of the court and have a duty to the court to give an account of the circumstances in which the will was executed (see Williams, Mortimer & Sunnucks 18th Ed. at para. 39-10).  And in a case like the present, where there is no legal requirement for there to be attesting witnesses and where no person who saw the deceased sign the alleged will is available to give evidence of that, a person in the position of Mrs Wang who had sworn an affirmation of testamentary scripts deposing to the circumstances in which the deceased gave the alleged will to her, telling that it was his will, becomes, in my opinion, a witness of the court.  The reasoning that treats attesting witnesses as witnesses of the court, amenable to being placed in the witness-box and cross-examined whether they like it or not, applies equally strongly to a witness in the position of Mrs Wang in the present case.  Mr Thomas QC submitted that Yam J had no power to direct her to go in the witness-box so that she could be cross-examined on her evidence about the provenance of the will.  I do not accept that that is right.  In my judgment, Yam J did have that power.  Mr Thomas QC said there was no judicial authority to that effect.  If that is right, I am content that this judgment should remedy the lack.  The power of the judge so to direct derives, in my opinion, from the inquisitorial character of probate proceedings in which it is the duty of the judge to try to give effect to the testamentary wishes of the deceased.  In order to do so in the present case, Yam J needed to investigate the provenance of the 1990 will.  He had, in the affirmations, the evidence of Mrs Wang about that.  If, as was plainly the case, he was suspicious about the truth of her evidence, he should have directed her into the witness-box to be cross-examined.  Moreover, it was open to counsel for Mr Wang senior to ask to cross-examine her on her affirmations.

644.  It was not Mrs Wang’s fault that the other side did not ask to cross-examine her or that the judge did not of his own motion direct that she make herself available for cross-examination.  I am easily persuaded that it did not occur to them that they could do so.  Mr Thomas QC described the way in which Mrs Wang’s affirmation evidence was dealt with as “a cock-up”, a description from which I do not dissent.  But that, too, was not Mrs Wang’s fault.  In the circumstances, however, it is not right to castigate her as being responsible for the fact that she was not cross-examined or to attribute to her a choice not to expose herself to cross-examination.  The only choice that it was in her power to make was a choice not to give any further evidence in chief.  That choice she certainly made. 

645.  So, what weight should the courts below have attributed to Mrs Wang’s affirmation evidence about the provenance of the envelope containing the 1990 documents?  Given the “cock-up”, as Mr Thomas QC described it, it would be wrong to treat the evidence as accepted by Mr Wang senior on the ground that it had not been challenged in cross-examination.  There are, plainly, a number of questions about it, never put and therefore never answered, that if put and answered might have clarified some puzzles.  But the evidence is consistent with, and therefore confirmatory of and confirmed by, Mr Tse’s evidence of the signing of the documents on 12 March 1990. 

The suspicious circumstances

646.  The conclusions to which I have come on Mr Tse’s evidence and Mrs Wang’s evidence conclude this case in Mrs Wang’s favour.  But I think I should not leave unchallenged some of the so-called suspicious circumstances relied on by Yam J, and the majority in the Court of Appeal, as tending to show that document A could not constitute the genuine will of Mr Wang. 

(1) It is said that Mr Wang had had no reason to change the intention, expressed in his 1968 will, to leave his whole estate to his father and nothing to his wife.  Yam J apparently thought that for Mr Wang to make a will in 1990 leaving everything to his wife and nothing to his father was so unreasonable as to be evidentially probative of forgery.  I do not hesitate to characterise this proposition as utterly ridiculous.  There seems no doubt but that in or around 1968 there were fairly serious marital difficulties between Mr and Mrs Wang and that Mr Wang suspected her, perhaps with reason, of infidelity.  But there is equally no doubt, on undisputed evidence adduced at trial, that by the early 1970s those difficulties had been put behind them and that in the fifteen years or so before the kidnapping in April 1990 they had together, as a team, built up a hugely valuable property based commercial empire and had consistently demonstrated the affection for one another and enjoyment of one another’s company indicative of a successful marriage.  The surprise is not that Mr Wang changed his will in 1990 making his wife rather than his father his sole beneficiary but that it had taken him so long to get round to doing so.

(2) Yam J treated it as suspicious that Mr Wang should have apparently made a home-made will.  But a home-made will made on 12 March 1990 was not, in my opinion, given the riding accident on 10 March, particularly odd.  It was not a serious accident but nor was it a trivial one as Mr Wang’s two days in hospital demonstrates.  It is not necessarily the severity of the injuries caused by a fall from a horse that may lead the rider to think about his or her will; it is the reminder of mortality that a fall, and particularly a fall that leads to a day or two in hospital, may produce that is apt to do that.  There are, however, undoubtedly two oddities associated with Mr Wang’s home-made will of 12 March 1990.  The first oddity is that, having made a home-made informal Chinese will on the day he discharged himself from hospital, he did not later make an appointment with his solicitors and have his home-made will transformed into a more formal state.  But his failure to do so cannot, in my opinion, reasonably be regarded as giving rise to a suspicion that the home-made will was forged.  It had taken some dozen years or more since his reconciliation with his wife for him to alter his out-of-date 1968 will.  This suggests to me a person who disliked thinking about his death or making arrangements for thereafter.

The second, and much more pronounced oddity, is that Mr Wang not only signed document A, the contents of which were obviously testamentary and suitable for a will, but also signed documents B, C and D and got Mr Tse to witness his signature on each of these documents as well.  The contents of documents B and C were precatory in character and, although somewhat inconsistent with one another, were related to the testamentary disposition to his wife that document A constituted.  But there was no need to have his signature on them witnessed.  They were merely instructions to his wife, who would surely have recognised his signature.  Document D, however, is an extraordinary document to find signed, witnessed and put in an envelope with documents A, B and C to be preserved until Mr Wang’s death.  Add to these oddities the oddity that the documents were in different handwriting (although B and C were written by the same person), not that either of Mr Wang or of Mrs Wang.  These oddities have not been, and in my view cannot be, satisfactorily explained.  But why are they indicative of forgery?  It is one thing to say that Mr Wang, or any rational testator, would not be likely to produce these documents, sign them all, have them all witnessed and then put them all together in an envelope for safe-keeping until his death.  But nor would any rational forger produce these four documents, unnecessarily quadruple the chances of the forgery being detected by forging the Wang signature on all four, and then put all four together in an envelope.  These oddities remain equally inexplicable whether the signatures are genuine or are forgeries and I do not think have probative weight of any significance pointing to forgery.

(3) Counsel for Mr Wang senior submitted to the judge that Mr Wang had had no opportunity to have these four documents prepared.  The judge accepted the submission.  I would not.  There is no indication when the contents of documents B, C and D were written.  They may have been written long before the riding accident.  And it would have taken no time at all for the contents of document A to have been written at Mr Wang’s dictation on 12 March 1990 either at the hospital before he discharged himself or in his office after he had discharged himself.  The judge described the questions of when, where and by whom the documents were written as “unanswerable and/or unexplainable questions unless the documents were forged”.  I think not.

(4) Yam J took the view that it would have been out of character for Mr Wang to have produced the four documents in the manner that, according to Mr Tse and Mrs Wang, he had done.  This is merely a repetition of the home-made will point with the added points that Mr Wang would not have had his will drafted in Chinese and would not have tolerated mistakes or alterations.  There is nothing in these points, in my opinion, reasonably probative of forgery.  Many testators do behave eccentrically. 

(5) Yam J commented on the sentiments expressed in the four documents and, particularly on those to be found in documents B and C.  He regarded the sentiments as inconsistent with Mr Wang’s feelings towards the persons concerned and, therefore, as indicative that the documents were Mr Wang’s documents.  The judge’s comments rehearsed some of the oddities to which I have already referred.  The features relied on as inconsistent with genuineness are equally incomprehensible if the signatures were forged. 

(6) Yam J put forward the absence of a revocation clause in document A as indicative of forgery.  The proposition that the omission in a home-made will of a revocation clause is indicative of forgery is as insubstantial as the “no reason to change intention” proposition (see (1) above).  It does not merit any further attention.

(7) Yam J took the view that Mrs Wang’s apparent knowledge that she was a beneficiary under document A was inconsistent with her evidence that she had not opened the envelope and was, therefore, suspicious.  It is true that in a number of affirmations that she had made before 6 September 1999 when the envelope containing the four documents was opened, and their contents were examined, Mrs Wang had referred to herself as “the beneficiary”, and on one occasion as the “sole beneficiary”, under the will that, she said, her husband had told her was contained in the envelope.  Counsel for Mr Wang represented these references as indicating that Mrs Wang must have known in advance of the envelope being opened what were the contents of document A.  There is, I think, a certain amount of weight that Mr Wang senior can derive from these references, but not very much.  An inference that she was a beneficiary under the will could reasonably have been drawn by her from the fact that her husband had put the will into her possession for safe-keeping.  And the reference to herself as “sole beneficiary” was made as part of her attempt to resist Mr Wang senior’s application for receivers of Mr Wang’s estate to be appointed.  Mrs Wang correctly anticipated that receivers, if appointed, would seek to challenge her control over the Chinachem empire.  She was anxious to combat the challenge.  I do not, for my part, find it a matter of surprise if, in attempting to do so, she may have over-egged the pudding in describing her believed interest in her husband’s estate.  I do not regard the references as constituting evidence of any significant weight in the forgery balance.  As to the prior knowledge that Mrs Wang is alleged to have had of the Lee Kin Sang signature (if that is who the second attesting witness was) the evidence was altogether too tenuous in my opinion to justify the judge’s conclusion that “she could not have known that unless she had already seen document A or otherwise was a party to the forgery of document A” (para. 5.106)

(8) Yam J treated as suspicious Mrs Wang’s behaviour in propounding the 1990 will.  For my part, I fail to understand why Mrs Wang’s reluctance to accept the fact of her husband’s death and her resistance to Mr Wang senior’s application to have his death presumed should be treated as a matter giving rise to suspicion.  It seems to me consistent with the reaction to be expected from a distraught lady whose husband had suffered an unknown fate and who was hoping against hope that the conclusion he was dead, a conclusion impelled by logic and probability, was not true.  I would, for my part, have been more inclined to be condemnatory about a father-in-law who, for the seven years following the kidnapping, had remained on ostensibly friendly terms with his son’s wife, perhaps his widow, but who, at the earliest moment made possible by the law, had instituted highly aggressive legal proceedings against her, characterising her opposition to his endeavours as based on forgery and fraud, allegations that were wholly speculative and made in order to wrest control from her of her kidnapped, and probably deceased, husband’s fortune.  I find the judgment of Yam J quite remarkable for its insensitivity to the effect on a grieving widow that her husband’s kidnapping and death and her father-in-law’s opportunistic aggression must have had.

647.  I regard the cumulative weight of the “suspicious circumstances” as negligible and quite inadequate to justify characterising Mrs Wang and Mr Tse as conspirators in a plan to propound a forged will. 

648.  I would therefore allow this appeal and admit document A to probate. 

Mr Justice Chan PJ:

649.  The Court therefore unanimously allows the appeal, sets aside the orders made in the courts below and orders that the will of Mr Wang Teh Huei dated 12 March 1990 contained in the document marked by Yam J as “Document A” be admitted to probate as his last will.

650.  The Court furthermore makes the directions relating to submissions as to costs and to ancillary directions contained in paragraph 457 of the Judgment of Mr Justice Ribeiro PJ, namely:

(a) directing that the parties be at liberty to make submissions in writing as to the costs of this appeal and of the proceedings below, any such written submissions on behalf of the appellant to be lodged with the Registrar of the Court within 28 days of the handing down of this judgment; any submissions in answer to be lodged on behalf of the respondent within 28 days thereafter; and any further submissions on the appellant’s behalf in reply within 21 days thereafter;

(b) directing that the parties be at liberty to apply in writing for any further directions bearing on the implementation of these orders; and

(c) directing that any oral hearing deemed necessary to resolve any procedural issues or issues ancillary thereto arising in connection with implementation of these orders be dealt with by a single Permanent Judge.

 

 

Mr Geoffrey Vos QC, Mr Benjamin Yu SC, Mr Jat Sew-tong SC and Mr Ramesh K Sujanani (instructed by Messrs Johnson Stokes & Master) for the appellant

Mr Neville Thomas QC, Mr Edward Chan SC, Mr Albert Tsang and Mr Victor Luk (instructed by Messrs K M Chan & Co) for the respondent

Table of Contents

Mr Justice Chan PJ:

Findings on handwriting by the lower courts

Effect of Court of Appeal’s finding re the Tse signatures 

Handwriting evidence not the only evidence 

Nature of handwriting expert evidence 

The Wife’s submissions on handwriting 

Principles of examination and comparison 

Knowledge of the Chinese language 

Features in Chinese handwriting 

Sample signatures used for comparison 

The Wang signatures generally 

Orientation and alignment 

General appearance and outlook 

Explanations for resemblance to the 1958 signatures 

Use of the calligraphic pen 

Naturalness 

Retouching in the writing - judge playing expert? 

Presence of tremors 

Significant similarities and dissimilarities 

Matching natural variations 

Inherent probabilities/improbabilities of forgery 

Consolidated evaluation 

Unfair criticism of Professor Jia 

Conclusion 

Mr Justice Ribeiro PJ:

A.    Introduction 

B.    The factual background 

C.    The proceedings prior to trial 

C.1    Application to swear death and other preliminary proceedings 

C.2    The appellant’s affirmations of testamentary scripts 

C.3    Opening the sealed envelope 

D.    The 1990 documents 

D.1    Document A – the 1990 Will 

D.2    Document B 

D.3    Document C 

D.4    Document D 

D.5    Features of the 1990 documents 

E.    The trial and the decision of the Court of Appeal 

F.    Concurrent findings and the present appeal 

G.    The legal principles 

G.1    The Wills Ordinance, Cap 30 

G.2    The burden of proof when a will is propounded 

G.3    The respondent’s case on forgery 

G.4    The relevance of Re H to the respondent’s case on forgery 

G.5    Drawing inferences of forgery or fraud

G.6    The respondent’s case involving Mr Tse and “the subtle form of forgery” 

G.7    Conclusions on burden and standard of proof summarised 

G.8    The presumption of regularity 

G.9    The second rule in Barry v Butlin 

H.    The approach of the courts below to the evidence 

H.1    Yam J’s approach to the factual evidence 

H.2    Yeung JA’s approach to the factual evidence 

H.3    Yuen JA’s approach to the factual evidence

H.4    The approach to reviewability on appeal

I.    The evidence relating to Mr Tse as attesting witness

I.1    The facts 

I.2    The respondent’s case on Mr Tse 

I.3(a)    Mr Tse as an improbable choice as attesting witness

I.3(b)    Mr Tse’s alleged denial that he knew he was witnessing a will 

I.3(c)    The statement given to Alex Lam as having been pre-prepared 

I.3(d)    The statement as being too well-written and mistake-free to be true 

I.3(e)    Inexplicably volunteered contents too well-tailored to the appellant’s case to be believable 

I.3(f)    Mrs Tse’s share in Global Cheer Limited 

I.3(g)    Failing to mention his witnessing documents to officers investigating kidnapping

I.3(h)    Mr Tse fleeing the jurisdiction

I.3(i)    Absence of impressions of Mr Wang’s signatures

I.4    The status of Mr Tse’s evidence

J.    The appellant’s affirmations of testamentary scripts

J.1    Whether they are evidence in the case

J.2    The weight to be given to the affirmations

K.    Attacks on the appellant’s credibility 

K.1    Delay in lodging the 1990 Will

K.2    The appellant’s illicit knowledge of the contents of the sealed envelope 

K.2(a)    The appellant’s affirmations 

K.2(b)    The appellant and Lee Kin Sang

K.3    Dr Kung and holiday wills

K.4    Failure to identify the author of the Lee signature

K.5    The absence of oral testimony from the appellant and the overall status of her evidence in the case

L.    The propensity of Mr Wang to make a will like the 1990 Will 

M.    Choice of the appellant as beneficiary

N.    Other “suspicious circumstances” 

N.1    Delay in revoking the 1968 Will

N.2    Why a homemade will?

N.3    Why such unconventional documents and idiosyncratic language? 

O.    The handwriting evidence

O.1    The nature of the expert handwriting evidence generally 

O.2    Features of the handwriting evidence in the present case 

O.3    Yam J’s approach to handwriting 

O.4    One example examined 

O.5    Abandonment of the Re H and Lee Ming Tee principles 

O.6    The Court of Appeal’s approach 

P.    Conclusions as to the evidence overall 

Q.    Wholesale copying and an unfair trial 

R.    Disposal of this appeal 

Mr Justice Litton NPJ:

Introduction 

Proof of the will

Mr Tse Ping Yim 

The questions for determination 

Whether the defendant could dispel suspicion

Inherent bias

“Subtle form of forgery” 

The Court of Appeal 

The weight of Mr Tse’s statements

Mrs Wang’s affirmation of testamentary scripts

Lee Kin Sang

Mrs Wang’s behaviour regarding the envelope

Mrs Wang’s failure to testify in open court

Handwriting

Matching variations 

Retouching

Calligraphic pen

Conclusion on Handwriting : Trial Court 

Treatment of Handwriting by Court of Appeal 

Pulling together all the strings 

Sir Noel Power NPJ:

Lord Scott of Foscote NPJ:

Introduction 

The issue 

The burden and standard of proof

Mr Tse’s evidence

Mrs Wang’s evidence

The suspicious circumstances

Mr Justice Chan PJ: