LINDA CHIH LING KOO v. LAM TAI HING

IN THE COURT OF APPEAL

1992, No. 116
(Civil)

________________

________________

J U D G M E N T

________________

Penlington, J.A.:

1. This is an appeal from a judgment of Bokhary J., as he then was, delivered on the 14th April 1992 following a trial lasting some nine weeks in which the respondents sued for alleged copyright infringement and breach of confidentiality on the part of the appellant in relation to a questionnaire which they had prepared and used in the course of medical research into the causes of lung cancer among non-smoking Chinese women in Hong Kong. He found that there was such infringement of copyright by the appellant in the way he had used the respondents' questionnaire in conducting his own research and that in so doing he had also used material which he knew to be confidential to the respondents. Bokhary J. also found that the appellant had obtained the respondents' questionnaire "surreptitiously", a word which was the subject of considerable debate during the hearing of this appeal. By agreement damages were to be determined by a Master, though Bokhary J. hoped that might not be necessary if the parties decided that they had had enough of expensive, time-consuming and clearly bitter litigation. It was a forlorn hope. The respondents were awarded the costs of the action, to be taxed on a common-fund basis, if not agreed.

2. The appellant's main grounds of appeal are that Bokhary J. was wrong to find that the appellant's and respondents' questionnaires were so "strikingly similar" that the appellant had infringed the respondents' copyright in their document. As regards the finding of breach of confidentiality the appellant says that the Judge was wrong to find that the respondents' questionnaire was confidential to them, it being compiled from previously published material, and that there was insufficient evidence to justify his finding that the appellant had obtained the respondents' questionnaire "surreptitiously". These were challenges to the Judge's finding of fact, based on his assessment of the evidence put before him and after a long trial. There can be no question that this placed a heavy burden on the appellant to show that not only that the Judge's findings were wrong but were such that if he had correctly applied the relevant legal principles he could not have reasonably come to the conclusions he did on the evidence before him.

Background

3. The background to this sad case is that in the 1970s and early 1980s research carried out in Hong Kong and elsewhere in South-east Asia had shown that there seemed to be an unduly high rate of lung cancer amongst Cantonese-speaking women. Other research had very clearly demonstrated that smoking was the pre-eminent cause of lung cancer but this high incidence also existed in women who did not themselves smoke. The discovery of the cause of this phenomenon was clearly a very worthy object of medical research and this was done by both the respondents and the appellant in Hong Kong.

4. The 2nd respondent, Prof. Ho, is an Honorary Professor of Radiation Oncology at the University of Hong Kong ("the University") and has been engaged in medical research for many years. He is a very senior figure at the University and enjoys the highest world-wide reputation. The 1st respondent, Dr. Koo is a lecturer at the University's Department of Community Medicine and has studied and is a graduate of Berkeley College in California. She obtained a doctorate from that college in 1976 in medical anthropology.

5. Dr. Lam, the appellant here, obtained an MBBS and MD in Hong Kong and two Master degrees from the University of London in Medical Sociology and Occupational Medicine. He is a Reader at the University.

6. The respondents commenced their research in 1980. The main tool of that work was a questionnaire by means of which it was hoped that a pattern would emerge indicating the cause, or likely cause, of the high rate of cancer in the target group of non-smoking women. The form of this questionnaire was completed in October 1980 and some preliminary testing with it was done in 1982 and 1983 by questioning 120 cases of women who had lung cancer and 120 "controls", women of the same racial background who had not contracted the disease. In 1984 the respondents prepared a report based on the findings which was published in 1985 in the "International Journal of Cancer" as "An Analysis of Some Risk Factors for Lung Cancer in Hong Kong". Further research, aided by modern computer technology has continued to the present time and a very large volume of material has resulted.

7. Dr. Lam became interested in this subject when he was in London in 1980-81 studying for one of his Master degrees. He completed a final research proposal in April 1982 and was granted funds to carry it out in July of that year. He then also turned to an important part of the research, the drafting of a questionnaire which would be most likely to result in correct conclusions being reached. This was completed by April 1983 and his field work then commenced. In 1986 he published a paper entitled "Smoking, Passive Smoking and Histological Type in Lung Cancer in Hong Kong Chinese Women". It was published in the "British Journal of Cancer" in 1987. One of the conclusions reached was that non-smoking women who were however married to husbands who smoked faced a higher risk of contracting lung cancer than ones whose husbands were also non-smokers. The respondents' conclusions had been that the cause of lung cancer amongst Chinese women who did not smoke was still elusive. Passive smoking did not show any significant increase in risk and other environmental factors were suspected. However in another paper published in the same journal in 1987, following further work, they concluded that passive smoking might well have some association with risk of lung cancer of a certain type.

8. Some of the other research studies which had been conducted were referred to in the course of the trial. The first was compiled by Dr. Robert MacLennan in Singapore in 1972-73. Dr. MacLennan was one of the expert witnesses called by the respondents. Dr. MacLennan assisted in a further study in Hong Kong in 1976/77 by Dr. W.C. Chan, Dr. Michael Colbourne, Mr. S.C. Fung and Prof. Ho himself ("the CCFH report"). That report looked into various factors which may contribute to the high and increasing rate of lung cancer occurring in non-smoking Chinese women, such as the use of kerosine or oil-based gas as a cooking fuel. No firm conclusion was reached and the report stated that further research was certainly called for.

9. Dr. Colbourne, one of the authors of the CCFH report, at the beginning of 1980 was Professor of Community Medicine at the University and was anxious that further study into this field of research be carried out. He approached the 1st respondent, Dr. Koo, who was a recent appointment to his staff, to continue research with Prof. Ho, and they agreed to do so. Dr. Colbourne himself however was then due to retire and later that year he went back to London when he shared a flat with the appellant and his wife. Dr. Colbourne was working on a paper to be presented in Edinburgh on lung cancer in Hong Kong women and the appellant contributed to the contents of that paper to such an extent that he was cited as a co-author when it was published in 1982 in Hong Kong. Dr. Colbourne gave expert evidence for the appellant and his successor, Dr. J.W.L. Kleevens, also testified as to matters of fact.

10. Another study into this subject was published in 1981 by Dr. Takeshi Hirayama in Japan. This was a very extensive project involving over 90,000 subjects who were non-smokers and took from 1966 to 1979. Dr. Hirayama came to some definite conclusions one of which was that such non-smoking women did run a higher risk of contracting lung cancer if they were married to heavy-smoking husbands.

11. The position in the early 1980s was then somewhat unusual. There were two teams of scientists at the University, which by international standards is not a large one, conducting research into the same phenomena. As one would hope and expect there was cooperation and mutual assistance between them but sadly that did not continue and it was the respondents' case that in any event such help did not include providing a copy of their questionnaire for the appellant to use in his field work.

12. It was the appellant's pleaded case that while he had seen a copy of the respondents' questionnaire he had made no use of it whatever. There was no question of his questionnaire being in any way derived from that of the respondents. It was, as he put it earlier in his evidence at the trial, "not his cup of tea". However during the course of being questioned by Bokhary J. that position dramatically changed and the appellant said that he may have made some limited use of the respondents' questionnaire. It is a matter of the greatest regret that that concession was made only towards the conclusion of a very long trial and not much earlier. While it still remained for the respondents to show not only that the appellant had derived his ideas for his questionnaire from theirs but he had actually copied it, the concession greatly weakened the appellant's case particularly on confidentiality in that it could no longer be argued on his behalf that even if the questionnaire was confidential, the appellant had made no use of it.

13. At the trial each side called two expert witnesses. For the respondents Dr. Robert MacLennan gave evidence as an expert cancer epidemiologist and Dr. Malcolm Coulthard as a linguist. Both have very high qualifications and considerable standing in their own fields, as do those called for the appellants.

14. For the appellant Mr. Richard Peto gave evidence generally about cancer research. He is Director of the Imperial Cancer Research Fund's Cancer Studies Unit at the University of Oxford and has been closely associated with Sir Richard Doll's pioneering work in the field. The appellant also called Dr. Michael Colbourne, head of the University's department to which the appellant and the respondents belonged from 1974 to 1980. He is an epidemiologist of very long experience. The appellant did not call any expert in linguistics.

15. The medical experts gave evidence as to the question of whether the appellant had copied the respondents' questionnaire, and as to whether such a document was confidential to its authors. They came to very different conclusions.

16. In the pleadings and during the course of the trial it was disputed whether copyright subsisted in the respondents' questionnaire and, if it did, whether each of them had ownership. However in his closing address Mr. Garland, counsel for the appellant at the trial, conceded that copyright did subsist in the questionnaire as a literary work and Bokhary J. found as a fact that the respondents were joint owners of that copyright. That finding is not challenged here.

Copyright

17. On the first question of whether the appellant infringed the respondents' copyright in their questionnaire in preparing his own the law is clear that while a person is entitled to use the ideas contained in another person's work, and it was eventually conceded that the appellant had done so, he must not reproduce the work or a substantial part of it. Furthermore in looking to see if the part reproduced is "substantial", it is proper to consider not only the quantity so copied but also the quality, i.e. was the part or parts copied an important part of the work even if it was a smaller part than the balance which was not so copied. Ladbroke (Football) Ltd. v. William Hill (Football) Ltd. [1964] 1 WLR 272 was a dispute over copyright in a football pools coupon. At 276-77 Lord Reid set out what he considered were the relevant principles as regards reproduction of another copyright work. He said:

" The Copyright Act, 1956, provides, by section 2, that copyright shall subsist in every original literary work and, by section 48, that literary work includes any written table or compilation. I have no doubt that the coupon must be treated as a single compilation. The appellants' dissection theory is derived from some statements in infringement cases and I must, therefore, examine at this point the law regarding infringement. Copyright gives the exclusive right to do certain things including 'reproducing the work in any material form' (section 2(5)), and reproduction includes reproduction of a substantial part of the work (section 49(1)). Broadly, reproduction means copying, and does not include cases where an author or compiler produces a substantially similar result by independent work without copying. And, if he does copy, the question whether he has copied a substantial part depends much more on the quality than on the quantity of what he has taken. One test may be whether the part which he has taken is novel or striking, or is merely a commonplace arrangement of ordinary words or well-known data. So it may sometimes be a convenient short cut to ask whether the part taken could by itself be the subject of copyright. But, in my view, that is only a short cut, and the more correct approach is first to determine whether the plaintiffs' work as a whole is 'original' and protected by copyright, and then to inquire whether the part taken by the defendant is substantial.

A wrong result can easily be reached if one begins by dissecting the plaintiffs' work and asking, could section A be the subject of copyright if it stood by itself, could section B be protected if it stood by itself, and so on. To my mind, it does not follow that, because the fragments taken separately would not be copyright, therefore the whole cannot be. Indeed, it has often been recognised that if sufficient skill and judgment have been exercised in devising the arrangements of the whole work, that can be an important or even decisive element in deciding whether the work as a whole is protected by copyright."

18. There can be no doubt that a degree of similarity between the two questionnaires was inevitable as they were both concerned with a relatively narrow field of medical research and the authors had each had access to the same previous publications by other scientists. Pike v. Nicholas 5 LR Ch 251 was a case concerning two books each written to secure a prize in ancient Welsh history. At 268 Gippard L.J. referred to the decision appealed from and cited with approval the words of James V.C. therein that:

"there is no monopoly in the main theory of the plaintiff or in the theories or speculations by which he has supported it, or even in the use of the published results of his own observations. It would, therefore, appear at once that the task undertaken by the plaintiff is an almost impossible one unless he can show there were substantial passages either actually copied or copied with mere colourable alteration."

19. On the central question of infringement of copyright Bokhary J. was in no doubt that there were substantial differences between the appellant's questionnaire (L4) and the respondents' (KH4d) which is quite clear from simply holding one in each hand. For instance, one can see from its format that KH4d (which was the final version of the respondents' questionnaire and the one used in their field research) was to be prepared entirely by asking the interviewee questions whereas the first two parts of L4 relied on obtaining, presumably with the interviewee's consent, data from hospital records. Both questionnaires go into the eating habits of the interviewee but whereas KH4d in section G "Dietary History" lists 15 foods or food groups, L4 in its section K while it also is headed "Dietary History" only lists nine, three of those being different sorts of cabbage. There are many other differences in form and content. Clearly Bokhary J. was well aware of these differences though he did not set them out in full. He did say:

" In no respect are the questionnaires identical. In every respect differences are, to a greater or lesser extent, to be found. All these differences must be weighed in the balance. But it is unnecessary and would greatly lengthen this judgment - to list all the differences which emerge from a comparison of the questionnaires. The differences have been stressed by Dr Lam's experts; and Dr Koo and Prof. Ho's experts have made no attempt to brush such differences aside.

Among the broad differences is this one. Dr Koo and Prof. Ho's questionnaire is essentially (although not exclusively) 'open-ended', while Dr Lam's questionnaire is essentially (although again not exclusively) 'close-ended'. Dr Lam's questionnaire probably gathered data in a form which required less effort to feed into a computer than the data gathered under Dr Koo and Prof. Ho's questionnaire.

Certainly, I do not regard the expression 'blind copying' - which is one used by Mr Fung in the course, I almost said 'the heat', of his cross-examination or Dr Lam - as apt, on any view, to describe Dr Lam's conduct.

But the matter does not by any manner of means end there, for 'blind copying' is hardly the definition of copyright infringement, let alone misuse of confidential information."

20. He then however went into the similarities between the questionnaires to see if they had the degree of similarity necessary to constitute infringement.

21. Dr. MacLennan prepared a table to show the similarities between KH4d, L4 and also previous published questionnaires, in particular that used for the CCFH report. He said that having compared the KH4d and L4 with the previous sources he found there were differences but several exceptional features were in both those of the respondents and the appellants. It was his opinion, based on careful analysis that it was extremely unlikely that the appellant had created his questionnaire independently of the respondents' one. It should however be remembered that that report was produced before the appellant admitted that he did make use of KH4d. It was the appellant's case here that Dr. MacLennan's evidence was as to derivation, not copying.

22. Dr. Coulthard gave evidence as an expert who had been used frequently to examine confession statements to see if they had originated from the suspect or from police officers. His report and evidence was referred to by Bokhary J. as follows:

" With the addition which he made while in the witness-box, Dr Coulthard's report concludes thus :-

'The encoding of ideas into language is a complex process:

Firstly, ideas are not linear, but text is. In other words there is often no logical or best sequence for presenting particular ideas or for asking particular questionnaire questions and on different occasions the same writer may well adopt a different sequence. As we have seen above the Questionnaires used for comparative purposes, 3-7 differ markedly from each other and from the Koo/Ho Questionnaire, while the Lam Questionnaire shows marked similarities to the Koo/Ho Questionnaire.

Secondly, on looking at both spoken and written texts we observe very clearly that, when asked to say or write the same thing on different occasions individuals encode differently. In other words when two texts have major similarities it is always because one has drawn on the other or because both have drawn on a third text. An examination of the earlier versions of the Koo/Ho Questionnaire shows some marked changes but one can always see that major portions have been copied verbatim from one version to the next.

Thus although Questionnaires KH4d and L4 are not in any sense versions of "the same questionnaire" the similarities between them are too substantial to have occurred by chance; there are only three ways of accounting for such similarities:

1) Questionnaire KH4d and its earlier versions may have incorporated items from (earlier versions of) L4. This, I gather, is highly unlikely, because Dr Lam began his research later.

2) Both the Koo/Ho and the Lam Questionnaires may have been based on another earlier questionnaire; in other words their similarities may be due to both having taken identical items from this third questionnaire. Of the other texts I examined Questionnaire 3 is the closest in content, organisation and wording to KH4d and L4 and I gather both teams had access to it; however, almost all of the similarities between the Koo/Ho and Lam Questionnaires are not to be found in questionnaire 3 so it is not a candidate.

3) Questionnaire L4 in part replicates (versions of) the Koo/Ho Questionnaire. This is the conclusion I must inevitably reach unless Dr Lam is able to produce another questionnaire on which he drew and on which Questionnaire KH4d can be demonstrated also to be ultimately based.'

What Dr Coulthard added in the witness-box are the words ' or because both have drawn on a third source'. The 'Questionnaires used for comparative purposes, 3-7' referred to by Dr Coulthard are, as he lists them in his report, these :-

'3. "Bronchial carcinoma study questionnaire", prepared by Professor Ho and colleagues for a 1976-77 study and to which both groups had access

4. English version of Shimizu et al's "Questionnaire for females in lung cancer epidemiology study" used in Japan in 1982

5. Translated version of questionnaire from lung cancer study in Guangzhou China, undated

6. "American Health Foundation comprehensive tobacco questionnaire", dated July 1986

7. Questionnaire for "International study on exposure to other people's smoke and urinary continine levels in non-smokers" in the design of which Dr Koo participated; the study was conducted in 1986'.

Number 3 is the questionnaire used in the CCFH study. In the course of the trial: number 4 was referred to as 'Shimizu's questionnaire'; number 5 was referred to as 'the Guanzhou questionnaire' or 'Du's questionnaire'; number 6 was referred to as 'the American Health questionnaire' or 'the AHF questionnaire'; and number 7 was referred to as 'Riboli's questionnaire'."

23. Mr. Peto was of a contrary view, and said that it would be normal practice to use other researchers' questionnaires in preparing ones' own. He considered however that KH4d and L4 were in any event not similar and the respondents' questionnaire had had little influence on the appellants. Even if it had that should not give rise to adverse comment.

24. Dr. Colbourne's conclusion in his report, as cited by Bokhary J., was that there was bound to be a degree of overlap in questionnaires on this same topic. Both included some common items but he could see no single phrase in the appellant's questionnaire which suggested that he had used the respondents' in framing it. The appellant's later evidence showed that that conclusion was not correct. Bokhary J. found that there were striking similarities between the questionnaires, a view shared by the linguistic expert Dr. Coulthard and also Dr. MacLennan. That he was entitled to be so guided by expert evidence was approved in Francis Day & Hunter Ltd. v. Bron and anr. [1963] 1 Ch 587 per Upjohn L.J. and Billhoper Maschinenfabrik Gmbh v. T.N. Dixon and Co. Ltd. [1990] FSR 105 per Hoffman J. (as he then was).

25. Bokhary J. found that there was a similarity in the sequence of questions which was more likely to be a reasoned choice than mere chance. KH4d begins with details of the patient and her illness before going on to demographic history. L4 does the same, though it provides for much more detail being obtained from hospital records. Each questionnaire then goes into the interviewee's smoking history and asks questions which are very similar about frequency of smoking, type of tobacco used and whether the smoke is inhaled. Each then goes on into detail about the patient's exposure to smoke from other persons at home and at work, though L4 goes into these matters in considerably more detail.

26. Then follows a section headed in KH4d "Occupational History" and in L4 "Occupational History and Smoking at Work". In KH4d the next section relates to "Personal Habits and Hygiene" and asks, inter alia, about the use of incense and mosquito coils (which are burnt to produce a smoke to repel insects). This information is also asked for in L4 but at the end of the questionnaire. However KH4d's next section is "Medical History" and Lo has a similar section "Family Medical History" and each goes into the causes of death of close relatives, and as is to be expected, in particular if they had cancer.

27. Then in both cases comes a section, in KH4d "Dietary History", in L4 "Cooking History". Questions are asked about what type of food is consumed, how frequently and how it is cooked.

28. Bokhary J. also found that the use of tables in both questionnaires supported a finding of similarity. Having gone through the use of such tables in both cases in detail he said:

" To summarise therefore: of the 8 tabular form questions in KH4 a to d, all but one are also to be found in L4 for cases and L4 for controls; and there are only three questions in that form in the L4 questionnaires which are not to be found in KH4 a to d. The position is the same when the comparison is made with KH3c save that in KH3c the question on passive smoking at home is in embryonic rather than fully-fledged tabular form."

29. It was clear from the evidence and the form of each questionnaire that the two most important avenues of investigation were first the inhalation of smoke and gases, whether they be from other persons, tobacco, incense, mosquito coils, cooking oil, kerosene or whatever. The second one was the consumption of food which possibly increased the risk of cancer, such as pickled fish or vegetables, or other food such as green vegetables which provided some degree of immunity, as had been found by Dr. Hirayama. As regards these important subjects Bokhary J. said this:

" What similarity is to be found here? As far as the less important inhalants are concerned, there are more dissimilarities than similarities. Dr Koo and Prof. Ho ask about many things about which Dr Lam does not ask, such as : living near power plants, incinerators and air-polluting industries; hobbies which may expose one to inhalants; cosmetics composed of fine dust; insecticides; mothballs, camphor etc. for the storage of clothing; and deodorizers and air fresheners. Dr Lam asks about cooking methods, while Dr Koo and Prof. Ho do not. None of that is to say that none of the less important inhalants are common to both questionnaires. As we have seen, both contain non-tabular questions on medicinal oils, incense and mosquito coils. As we have also seen, both gather information in tabular form on these two matters, which seem fairly important in regard to inhalation even though less important than passive smoking : exposure to dust, chemicals and the like at work over the years; and cooking fuel use over the years."

As regards ingestants he found that:

" Turning to ingestants, the most significant similarity as to form between the two questionnaires lies in this fact. Each has a list of specific food items; each provides a frequency scale for measuring the consumption of such food items; and the scales are quite alike."

30. Some reliance to show copying by the appellant was the use of phrases common to both questionnaires which were unusual. The first is a grammatical mistake in both documents. KH4d under "Patients Demographic Background" asks about "age of marriage". In 6.4 under "Demographic History" the question is "age of (first) marriage" (my emphasis). Bokhary J. accepted that while Dr. Koo's English is generally excellent, as one would expect, she does consistently make this mistake of putting "of" instead of "at" in this sort of context. She did it elsewhere. The appellant's evidence on this was that it was a common mistake amongst Hong Kong Chinese. He was however referred to other documents of which he was the author or co-author in which the correct use of "at" is found. The suggestion was that he only makes the mistake when led into error by copying others.

31. The other phrase was "pork fat", instead of the more common word "lard", being used in KH4d in "Dietary History" and in L4 under "Cooking History". The appellant said the use of the words "pork fat" was deliberate as he was interested in the use of actual raw fat which first had to be rendered before the oil could be used for cooking. This may have given off fumes which were harmful. That explanation, however, should be looked at in the context of the way "pork fat" is placed together with other products which are all oils under the heading "Cooking oil". Moreover it was accepted by the appellant that the Chinese character beside the words "pork fat" in L4 means "pork-oil" which would not fit in with what he says his interest was. Bokhary J. concluded that if he took an unfavourable view of the appellant's credibility these two instances were straws in the wind pointing to the copying of KH4d in L4. Bokhary J. later did come to that view of the appellant's evidence.

32. It was part of the defence to this claim, both on copyright infringement and breach of confidentiality, that the respondents' questionnaire did not contain information which was not freely available in the public domain. Previous research, and some of the questionnaires on which it was based had been published and any competent researcher could, with that information, have compiled the questionnaire.

33. The respondents rely on Elanco Products Ltd. v. Mandops (Agrochemical Specialists) Ltd. [1979] FSR 46, which was an appeal against an interim injunction relating to the use of a leaflet accompanying cans of a herbicide. The information in the leaflet was available but it was held that the question to be answered was: would the defendants by using the literature they were anxious to use, be making use of the skill and judgment of the plaintiffs or would they have done their work for themselves? The defendants were fully entitled to make use of any material which was available to them as being in the public domain for the purpose of compiling their literature. However they were not entitled to copy the plaintiffs' trade literature thereby making use of the plaintiffs' skill and judgment and saving themselves the cost of assembling their information and avoiding making their own selection of material to put into that literature. At p. 57 Buckley L.J. said:

"I agree. As I understand the law in this case, the defendants were fully entitled to make use of any information, of a technical or any other kind, which was available to them in the public domain, for the purpose of compiling their label and their trade literature, but they were not entitled to copy the plaintiffs' label or trade literature thereby making use of the plaintiffs' skill and judgment and saving themselves the trouble, and very possibly the cost, of assembling their own information, either from their own researches or from sources available in documents in the public domain, and thereby making their own selection of material to put into that literature and producing their own label and trade literature."

34. Leading counsel for the appellant, Mr. John Griffiths Q.C., who was not counsel at the trial, has argued that Elanco Products Ltd. v. Mandops (Agrochemical Specialists) Ltd. should be looked at on the basis that it was not determinative of the issue, being an appeal from the grant of an interlocutory injunction, and all the Court of Appeal was deciding was whether there was a triable issue and where lay the balance of convenience.

35. So far as legal principles are concerned, I would not treat Elanco Products Ltd. v. Mandops (Agrochemical Specialists) Ltd. in the way that has been suggested. I find it of considerable assistance as to the way in which a document prepared using skill and judgment, which clearly was the case with the respondents' questionnaire, should be regarded even if the material from which it was drawn was in the public domain. There was evidence, including that of Dr. Kleevens, that the questionnaire was the "cutting edge" of this sort of research. It was vital that the correct questions should be asked so as to maximise the chance of finding the elusive cause of lung cancer in non-smoking women. This was, however, rendered more difficult because the interviewees would often be ill-educated and were moreover mostly suffering from a terminal disease. While the questions had to cover all reasonable possibilities it should not be such as to cause unnecessary distress. Because of this Dr. Kleevens said he insisted that the respondents' interviews be concluded before the appellant began though this did pose problems with funding and, from the beginning of 1983, caused friction in the department.

36. I am satisfied that the evidence showed that while the information required to prepare KH4d was generally available, the respondents did exercise a very considerable amount of skill and effort in its compilation and they were entitled to protection from copying without permission.

37. This is an appeal from a finding of fact by Bokhary J. and Elanco Products Ltd. v. Mandops (Agrochemical Specialists) Ltd. seems to me to say that if the trial judge finds that the alleged copyright infringer has made revisions to the material he had copied he may go on to find that he has not infringed the copyright, but he may not do so. It is for him to see if sufficient was done by the alleged infringer's own efforts to avoid liability (per Goff L.J at p. 55 and Buckley L.J. at 57).

38. As regards the question of onus of proof reliance is placed by the appellant on dicta of Upjohn L.J. in Francis Day & Hunter Ltd. v. Bron and anr. [1963] 1 Ch 587 at p. 618. There are, however, words in that passage which support the respondents, this being, as I have said, an appeal from findings of fact. Upjohn L.J. said this:

" This is really a question of fact and nothing else, which depends on the circumstances of each case; but it is a question of fact which must be taken in two stages. The first stage is objective, and the second stage subjective. The first question is whether in fact the alleged infringing work - which for the sake of brevity I will inaccurately call the defendant's work, for though the composer was a witness, he was not a defendant - is similar to the work of the original author - which again for the sake of brevity I will, with equal inaccuracy, call the plaintiff's work. Is it then proper to draw the inference that the defendant's work may have been copied from the plaintiffs' work? This is purely an objective question of fact, and depends in large degree upon the aural perception of the judge, but also upon the expert evidence tendered to him; but it is essentially a jury question. A defendant might in theory go into the witness-box and say that he had deliberately made use of the plaintiff's work, but that it is not an infringement, either because he did not make use of a substantial part of the plaintiff's work, or that, though the plaintiff's work has been utilised, he has been able so to alter it that it cannot properly be described as a reproduction. The onus is on the plaintiff to prove the contrary as a matter of purely objective fact, and if he cannot do so then the morally dishonest defendant will escape the consequence of the allegation of infringement. No such question arises in this case. At this stage similarity has been found by Wilberforce J., and that is not challenged before us. For myself, I think that perhaps I would have used rather stronger adjectives than 'definite' or 'considerable' similarity, which were the words used by the judge; the adjective 'close' would be more appropriate, but nothing, I think, turns upon that matter." (the appellant relies on the words emphasized)

39. Here the appellant has conceded that he did make use of the respondents' work but says also that it was not a substantial use and his questionnaire is so different in form from that of the respondents that there is no infringement. The onus was on the respondents to prove to the contrary. Upjohn L.J. however also says that is a purely objective question of fact. It is a "jury question" and depends on the judge's perception and the expert evidence. Here there is nothing to show that the judge was not aware of that and he has very carefully gone into the reasons for his finding and has availed himself of the expert evidence, in particular that of Dr. MacLennan and Dr. Coulthard. He also considered the evidence of the appellant himself and of the respondents. He accepted theirs but not that of the appellant.

40. Reliance is also placed on the earlier words of Willmer L.J. at 614 when he approved what had been submitted for the plaintiffs:

"Where there is a substantial degree of objective similarity this of itself will afford prima facie evidence to show a casual connection between the plaintiff's and the defendant's work; at least it is a circumstance from which the inference may be drawn."

I would accept that this does not mean that if derivation is proved, or as here conceded, that casts a burden on the defendant to disprove objective similarity but I do not see anything in the judgment to indicate that Bokhary J. took that view.

41. Willmer L.J. went on at 615 as regards a finding at first instance of objective similarity:

" The question, being one of fact, is eminently one for the determination of the trial judge, as I think is recognised in the American cases which were cited to us. It is to be remembered that Wilberforce J. not only had the advantage, denied to us, of himself seeing and hearing the witnesses at first hand; he also had the advantage, which strikes me as being of great importance, of hearing how the musical experts who were called as witnesses illustrated the technical evidence which they gave by demonstrations, both vocally and on the piano. Bearing this in mind, and having regard to the judge's acceptance of the evidence given by de Angelis, it is, in my judgment, impossible for us in this court to say that he reached a wrong conclusion on what as eminently a question of fact for him."

That dicta seems very apt to the considerations here.

42. It is clear that there are substantial dissimilarities between the two questionnaires, and Bokhary J. was fully aware of that, as anybody looking at them must have been. It is however accepted by counsel for the appellant that that part of the respondents' work which is alleged to have been copied must to be compared with the relevant parts of the appellants and dissimilarities in other parts are irrelevant to that task. Dissimilarities in those parts being compared are of course relevant but I think it is clear from the judgment that Bokhary J. did have regard to such dissimilarities.

43. Section 1(1) of the Copyright Act 1956 (extended to Hong Kong by the Copyright [Hong Kong] Orders in Council 1972 and 1979) gives the owner of the copyright the exclusive right to do or authorise the acts set out in relation to the work. Others may not copy it or a substantial portion of it without consent but it is not the ideas which may be the genesis of the work which are protected but the expression of those ideas; the form which that expression takes. At first sight it is perhaps difficult to see from merely looking at the questionnaires the close similarity which is required to found a claim for breach of copyright but Bokhary J. has very meticulously indeed gone into the relevant parts of the questionnaires, and, with the aid of the expert evidence, has found much similarity.

44. The principles upon which this Court should be prepared to upset such a finding of fact are very clearly and fully set out in the dicta of Fuad J.A. in The Hong Kong and Shanghai Banking Corporation v. Chan Yiu-wah and anr. [1988] 1 HKLR 457 at p. 474-477. There he sets out all the relevant authorities such as s.s. Houtestroom v. s.s. Sagaparack [1927] AC 37 (Lord Summer's dicta at p. 47) and Yuill v. Yuill [1945] P 15. Having done so he then considered the evidence in that case in detail and in several very important aspects found that the trial judge in coming to his findings of fact had not tested the demeanour of some witnesses against the whole of the evidence in question, as Lord Greene in Yuill v. Yuill said he should. For that reason he would have allowed the appeal. Both Hunter and Kempster JJ.A. took an opposite view but did not disagree in any way with the principles set out by Fuad J.A.

45. Here, on the question of breach of copyright in particular, the oral testimony of the appellant and respondents themselves were perhaps not as important as in The Hongkong and Shanghai Banking Corporation v. Chan Yiu-wah but Bokhary J. did come to an assessment of their credibility which clearly was of great importance as regards breach of confidentiality. He said:

" Dr Koo and Prof. Ho impressed me as honest and reliable. Their answers were always straightforward. They were never evasive.

Dr Kleevens did not impress me as reliable on matters of controversy. From the things which he said and the way in which he said them, I formed the view that his memory of those matters is unreliable. In respect of some of them he admitted as much. And even where he asserted a confident recollection, his demeanour was suggestive of far less confidence than his words alone would convey.

For a professional man, or indeed anyone, who is being sued and against whom serious allegations are made, giving evidence, especially in the course of a long and searching cross-examination, is likely to be a great strain whatever there may or may not be in those allegation. Due allowance for such strain must be made in Dr Lam's favour. But even having done that throughout, I am, having seen and heard him in the witness-box, left with an unfavourable impression of Dr Lam as a witness. My impression of him is the opposite of my impression of Dr Koo and Prof. Ho. I wish that I could - but I do not think that I can - avoid saying this : I do not think that Dr Lam's testimony was full or frank."

46. So far as the experts were concerned Bokhary J. found that all did their best to assist him, and each has, to some extent, enabled him to see things more clearly.

47. The appellant has mounted a strong and eloquent argument that Bokhary's finding of objective similarity between the two questionnaires cannot be supported and is plainly wrong. At first sight, simply looking at the questionnaires, there seems to be much in that argument - as Bokhary J. himself has said the dissimilarities are more obvious than the similarities. I have had the advantage of reading in draft the judgment of Nazareth J.A. and I too shared his initial reaction to the appearance of the two questionnaires. However Bokhary J. clearly has assessed the credibility of the witnesses as to fact in the way suggested by Lord Greene in Yuill v. Yuill. He has carefully considered in detail the written material that was before him and has been assisted by expert witnesses of the highest calibre in coming to his decision on objective similarity and I am myself unable to say that he was plainly wrong to come to the conclusion that he has.

Misuse of Confidential Information

48. This is a field which has been widened in recent years but the test as to whether there has been a misuse of confidential information still rests in the decision of Megarry J. (as he then was) in Coco v. A.N. Clark (Engineers) Ltd. [1969] RPC 41, a decision which followed Saltman Engineering Co. Ltd. v. Campbell Engineering Co. Ltd. [1948] 65 RPC 203 and Seager v. Copydex Ltd. [1967] RPC 349. It was approved in Fraser v. Thames Television Ltd. [1984] QB 44 where the authorities are comprehensively reviewed by Hirst J.

49. Megarry J. held that three elements had to be proved for a claim to succeed:

(a) that the information was of a confidential nature;

(b) that it was communicated in circumstances importing an obligation of confidence; and

(c) that there was an unauthorised use of the information.

That decision was cited by Bokhary J. and he attempted to follow it.

50. Here it seems clear that the respondents were on much stronger ground as soon as the appellant acknowledged in reply to a question from Bokhary J. that had he had made some use of KH4d. This is particularly so in light of the very late stage in a very long trial when it was made.

51. Mr. Griffiths Q.C. asked the rhetorical question in relation to the appellant's possession of KH4d "If Dr. Lam had obtained it unlawfully why would he have disclosed to Dr. Koo that he had it and produced it? It would have been so easy to simply destroy it." Equally, however, one could ask "Why, if Dr. Lam had in fact made some use of KH4d did he not simply agree right from the start that he had, make some acknowledgement of its use which would have satisfied the respondents and thus saved the enormous cost in time and money of the trial." There is probably no satisfactory answer to either question.

52. The evidence as regards confidentiality is that Dr. Koo said that she and Prof. Ho had worked on many versions of a questionnaire which was to form the basis for their research. She said, and this was supported by the experts, the questionnaire was the most important tool to be used in this sort of research and the validity of the findings depended on it. She was conscious that the persons being interviewed would mainly be poorly educated and there would be considerable difficulty in obtaining accurate and full information about a problem to which there seemed no clear answer. It involved going into past history such as previous eating habits, as lung condition was influenced not only by what was inhaled but also by ingestants and environmental contacts.

53. The work started in January 1980, though she and Prof. Ho had had many meeting before that. She consulted a mass of written material. She consulted other researchers and continued having meetings with Prof. Ho. She applied for and was granted a research assistant, Miss Nancy Lee. She gave a very great deal of evidence, the relevance of which seems rather peripheral, as to the progress that was made, the consultations and discussions with others in the field both in Hong Kong and overseas and clearly from her evidence, which was accepted, a very great deal of work went into the questionnaire which finally emerged as KH4d.

54. There was evidence, accepted by Bokhary J., that a very great deal of work and skill had gone into the compilation of KH4d, the last of many versions. Even if the material necessary to draft the questionnaire was in the public domain, if authors used skill and produced something novel from that material, that written material is confidential to them. Support for that is found in Interprint Comparison (Australia) Ltd. v. Law Society of New South Wales [1977] RPC 137 per Bowen J. who cites the dicta of Lord Greene M.R. in Saltman Engineering Co. Ltd. v. Campbell Engineering Co. Ltd. at 415:

"the information must be confidential and to be confidential, must, I apprehend, apart from contract have the necessary quality of confidence about it namely it must not be something which is public property and public knowledge. On the other hand it is perfectly possible to have a confidential document, be it a formula, a plan, sketch or something of that kind which is the result of work being done by the maker upon materials which may be available for the use of anybody; but what makes it confidential is the fact that the maker of the document has used his brain and thus produced a result which can only be produced by somebody who goes through the same process".

55. Bokhary J. also considered other authorities beginning with Prince Albert v. Strange. There are two reports at 64 ER 293 and 41 ER 1171. That concerned the unauthorised use of drawings and etchings made for Queen Victoria and Prince Albert of their family. It was held that such unpublished work could not be published and the courts would intervene to prevent it by granting an injunction. Bokhary J. considered this was authority for the proposition that there was proprietary interest in confidential information even though some doubts were cast on that in A.G. v. Guardian Newspapers (no.2) [1990] AC 109 - the "Spycatcher Case". Be that as it may Bokhary J. was in no doubt, and it was not disputed at the trial, that the courts have power to intervene to protect the unauthorised use of confidential information either imparted in circumstances showing that it was confidential or obtained without authority, as is the case here.

56. As regards the appellant Dr. Koo said she first met him in 1980 and saw him again in Hong Kong at the University in mid-1981. He started his own research and she continued with hers. Relations were cordial but that situation deteriorated in 1984. In 1983 Dr. Kleevens, then head of the Department, had suggested to Prof. Ho that the appellant should join his and Dr. Koo's research project. In a letter to Dr. Kleevens Prof. Ho declined and Dr. Koo said this created some problems. She said in fact that Dr. Kleevens wanted her to finish her project as soon as possible and there would be no further funding. At one stage Dr. Kleevens was on leave and the appellant, who had enjoyed swift promotion, was acting Head of the Department, and there were acrimonious exchanges between them. This attitude extended to Dr. Koo's assistants and clerks. Dr. Koo then went abroad to attend a course at the National Cancer Institute in the USA but on her return relations did not improve. This evidence was supported by Dr. Kleevens who said that he wanted the respondents to finish their field work before the appellant started and due to consequent funding problems relations were not good in early 1983.

57. Dr. Koo and the appellant worked on the same floor and their research assistants shared an office and in March 1986 it was Dr. Koo's evidence that she had reason to suspect that the appellant was in possession of a copy of KH4d. Relations between them at this time were not good. She said she went to see him and asked if he had a copy of her questionnaire. He replied that he thought he did but she had given it to him, a claim which she has consistently denied. Following that the appellant came to Dr. Koo's office and said he did have a copy of her research proposals with an "interview pattern" annexed, some other papers and a copy of her questionnaire. It was Dr. Koo's evidence, accepted by Bokhary J. that she had given the research proposal and interview pattern and the other papers to Dr. Kleevens as they were necessary to secure funding for the research. She was adamant that she did not however give Dr. Kleevens, or the appellant, the questionnaire, KH4d, or any of its predecessors.

58. Following this meeting the appellant came to Dr. Koo's office on the 24th and a further discussion took place. What was said was in dispute.

59. On the 29th March Dr. Koo wrote to the appellant saying that he had come to her office on the 24th and apologised for having a copy of her questionnaire. She denied having given it to him or to anybody else at the University. She said she was taking legal advice and reserved her right to pursue the matter. To this the appellant replied saying that all the papers, including the questionnaire, had been given to him by Dr. Koo some years before when he discussed his plans with her at the suggestion of Dr. Kleevens. He said that in designing his questionnaire he did not refer to hers at all, an attitude maintained until towards the end of the trial some years later. Finally there was a meeting of the parties in the office of the Vice-Chancellor in May 1986 but, very unfortunately, the dispute was not resolved. The appellant continued to deny making any use at all of KH4d.

60. Dr. Kleevens gave evidence that he thought Dr. Koo had given him an earlier version of her questionnaire because he recollected discussing with her the final question in it relating to washing rice. However he later said he had no firm recollection of being given that questionnaire, KH3d, and Bokhary J. expressly said he did not accept Dr. Kleevens' evidence about the conversation. That is not challenged.

61. Bokhary J. also considered the suggestion that the questionnaire may have been given to the appellant via his and Dr. Koo's respective research assistants, Miss S.C. Wong and Miss Nancy Lee. He found no evidence to support that theory (neither Wong or Lee was called) and found that in any event Miss Lee had no authority to give a copy of the questionnaire to Miss Wong. He found that only the respondents themselves had a copy of their questionnaire.

62. That the appellant knowingly made use of information which was confidential finds considerable support from evidence that not only was the appellant in possession of KH4d but he had circled two very important questions on his copy. These were questions C12 and C13 relating to smoking by other persons at home and at the interviewee's workplace. This seems clear evidence that the appellant considered this aspect - passive smoking at home and at work - was important and he said so in evidence. If that was so his reluctance to admit making any use whatever of KH4d lends support to a finding that he realised the questionnaire was confidential and he had no right to use it in the way he did.

63. Bokhary J. found that the appellant obtained a copy of KH4d "surreptitiously" a word which has been used in the judgments in other cases. It is defined in the Oxford Dictionary as "secret and unauthorised; clandestine". If this is to be taken as the equivalent to a finding that the appellant stole the document from Dr. Koo or her assistant it cannot, in my view, be supported by the evidence and Mr. Daniel Fung Q.C., leading counsel for the respondents here and at the trial, does not contend that it does. Dr. Koo said only she and Prof. Ho had copies and that evidence was accepted, though if that excluded Dr. Koo's own research assistant Nancy Lee it seems surprising. The appellant said that he had been given a copy of the questionnaire when he was given a copy of the respondents research proposal and interview pattern. It was at a time when relationships were friendly. Bokhary J. has found as a fact that that was not so but was unable to say how the appellant did obtain the questionnaire. He was entitled to find a surreptitious taking only to that extent - that the appellant did somehow come into possession of the document, and he must have known it was confidential because of the amount of work which had gone into its preparation. It had not been given to him by the persons whose information it was and again he must have realised he was not entitled to use it.

64. The appellant is a professional medical researcher with the highest qualifications and clearly has been highly regarded by Dr. Colbourne and Dr. Kleevens. There was no relevant evidence as to how the appellant came to be in possession of KH4d apart from Dr. Koo denying she had ever given him a copy, which was accepted, and the appellant saying she had done so when relations between them were cordial. There was however quite a lot of evidence which showed that Nancy Lee and Miss S.C. Wong and other research assistants shared office accommodation and it would not be in the least surprising that papers relating to one team should be mixed with those of the other. Any finding which goes beyond the appellant realising he was in some way in possession of KH4d, that it was confidential and that he was not authorised to make use of it, is not justified by the evidence.

65. For the claim on breach of confidentiality to succeed it was necessary for the judge to find that the second of Megarry J.'s conditions - the material was obtained in circumstances importing an obligation of confidence - had been satisfied. He clearly was so satisfied that no matter how the appellant obtained his copy of KH4d he must have realised it was not available for him to use.

66. There is no dispute that the third condition was met as the appellant, albeit reluctantly, admitted had he did make some use of the respondents' questionnaire. The appellant's counsel at the trial, and junior to Mr. Griffiths Q.C. here, Mr. Garland, conceded that while the appellant had not copied the questionnaire he had used "some ideas".

67. We were referred to authorities for the proposition that not only must a person in receipt of confidential information use it directly, he must not use it as a "spring-board" to assist in his own work.

68. In Seager v. Copydex Ltd [1967] 1 WLR 923 at 931 Lord Denning refers the decisions of Roxburgh J. in Terrafin Ltd v. Builders Supply Co. (Hayes) Ltd. [1960] RPC 128 and Roskill J. in Loanheigh Precision Engineering Ltd. v. Boyant [1963] 1 WLR 1293. In that later case Roskill J. said:

"As I understand it, the essence of this branch of the law, whatever the origin of it may be, is that a person who has obtained information in confidence is not allowed to use it as a spring-board for activities detrimental to the person who made the confidential communication, and spring-board it remains even when all the features have been published or can be ascertained by actual inspection by any member of the public."

Lord Denning went on:

"When the information is mixed, partly public and partly private, then the recipient must take special care to use only the material which is in the public domain. He should go to the public source and get it: or at any rate, not be in a better position than if he had gone to the public source. He should not get a start over others by using information which he received in confidence."

69. This "spring-board" concept is not without difficulty but it does mean that even if the appellant did go to other sources apart from the respondents' questionnaire, and it seems clear he did so, he was still not entitled to use it even in a limited way. However there was here evidence, which was accepted, of substantial use being made.

70. There is a possible fourth condition required to found a claim for breach of confidence, detriment to the other party, though there is authority to the contrary. Bowen J. refers to it being a requirement in Megarry J.'s judgment in Coco v. A.N. Clark (Engineers) Ltd. though Megarry J. at p. 48 he is clearly in some doubt. He said:

" Thirdly, there must be an unauthorised use of the information to the detriment of the person communicating it. Some of the statements of principle in the cases omit any mention of detriment; other include it. At first sight, it seems that detriment ought to be present if equity is to be induced to intervene; but I can conceive of cases where a plaintiff might have substantial motives for seeking the aid of equity and yet suffer nothing which could fairly be called detriment to him, as when the confidential information shows him in a favourable light but gravely injures some relation or friend of his whom he wishes to protect. The point does not arise for decision in this case, for detriment to the plaintiff plainly exists. I need therefore say no more than that although for the purposes of this case I have stated the propositions in the stricter form. I wish to keep open the possibility of the true proposition being that in the wider form."

Here I am satisfied that the evidence of Dr. Koo, accepted by Bokhary J. was that the use by the appellant of material from KH4d did cause doubts to be cast on the originality of the respondents' work and that clearly was detrimental to them.

71. I would also dismiss the appeal against Bokhary J.'s finding on breach of confidentiality.

72. I would only finally pay tribute to the meticulous way this appeal has been prepared and argued on both sides and I mean no disrespect by not referring to all the authorities which were cited to us.

Nazareth, J.A.:

73. I agree.

74. I would only add the following. At the very heart of the breach of copyright claim was the issue of substantial similarity between the two questionnaires. I feel bound to say that the first and strong impression from my own visual comparison of the two questionnaires, and reinforced by Mr Griffiths' powerful submissions, was of dissimilarity rather than similarity. However, the judge clearly was aware of and did not overlook such dissimilarity. But, as he was entitled to do, he accepted the evidence of the respondent's experts, and was impressed by their views, in particular of similarity in the specialised context of epidemiology. I am satisfied that it is not possible for this Court to fault the conclusion the judge reached, of substantial similarity.

75. Turning to the only other matter which I think does call for comment in this unhappy case, the reference to "surreptitious" taking was I think unfortunate. I am not persuaded that it was intended in the pejorative sense complained of by the appellant, and the evidence does not in my view support such a conclusion. On the contrary, with the respective staff and assistants of the two parties working in proximity and in broad terms, within the same organisation, it seems to me most likely that the respondent's questionnaire might have come into the possession of one of Dr Lam's assistants in some manner that was not surreptitious. That, of course, does not affect the conclusions my Lord has reached, with which, as I have indicated, I agree.

Sears, J.:

76. At the heart of this bitter and protracted litigation lies the inability of Dr. Lam to recognise that he utilised Dr. Koo's questionnaire in the preparation of his own. It may be that the lapse of time has convinced him that he has done nothing wrong.

77. The trial lasted some 9 weeks and every conceivable point which could be raised was. What appeared to be the resolution of simple issues of fact turned into a complicated medico/legal struggle with no quarter asked nor given. The main cause of this costly exercise was Dr. Lam and his lawyers failing to accept that Dr. Koo was entitled to copyright protection in respect of her questionnaire and that Dr. Lam had used it. Indeed it was not until the 40th day that Counsel conceded that copyright subsisted in the questionnaire as a literary work.

78. The Judge came to a clear conclusion on the evidence placed before him that Dr. Lam was not a truthful witness, that he had by copying substantial parts of Dr. Koo's questionnaire infringed her copyright, and that, as it was a confidential document, by obtaining and using it in an unauthorised manner was in breach of that confidence.

79. These findings, although reached after a mass of documentary and oral evidence had been produced, are essentially straightforward findings of fact. The appeal by the defendant concentrates on two grounds:

1. The Judge erred in finding there was substantial similarity between the two questionnaires.

2. The Judge, in the confidence claim, should not have found that the portions of the document which Dr. Koo claimed as confidential were and that Dr. Lam "surreptitiously" obtained it.

80. I must confess that when I first visually compared the two questionnaires, they appeared dissimilar. However, that approach is superficial and fails to have regard to the expert evidence that, as the Judge found, a questionnaire on lung cancer is an important tool in epidemiological research and as Dr. MacLennan put it "The questionnaire is the key instrument for data collection in studies such as those under consideration and its structure and content reflect the particular approach of an investigator".

81. It was therefore important for the Judge not to compare this document on a visual or quantitative basis, but to have regard to features which, to a layman, may not have appeared significant. The expert evidence called by the plaintiffs - Dr. MacLennan, a cancer epidemiologist and Dr. Coulthard, a linguist - demonstrated that there were a number of striking similarities between the two questionnaires which indicated that Dr. Lam had derived his from the other and that, objectively viewed, Dr. Lam's was similar to a substantial part of Dr. Koo's.

82. The Judge analysed all the evidence and found that Dr. Lam, as a matter of subjective fact, had derived his questionnaire from Dr. Koo's. There is no appeal against that finding.

83. It must be remembered that although copyright attaches to the questionnaire, all that the plaintiff has to prove is that the defendant has copied a substantial part of the questionnaire - see Francis Day and Hunter Ltd. v. Bron [1963] 1 Ch. 587 and that this substantial part is measured not in quantity but in quality - see Ladbrokes v. William Hill [1964] 1 WLR 273.

84. There are Number of dis-similarities between the two questionnaires and, as the subject matter of the two was the same, there will obviously be similarities. The Judge, in a careful analysis, aided as he was by the experts, set out what he considered to be the striking similarities. The two important research factors were the effect of passive smoking and ingestants as potential causes for lung cancer in non-smoking women.

85. The Judge considered that the treatment of those two factors was strikingly similar. I confess I have not found it easy, with a layman's eye, to recognise all these striking similarities, but this Court is not re-trying the action, nor should it give the impression that it is. Any appeal on fact is very difficult to mount and with a 9 week trial, where it is clear the Judge took the greatest care in analysing the evidence, the appellant really has to show either that there was no evidence to support the Judge's findings, or that he was plainly wrong. This appeal has itself lasted 12 days and the Court has been subjected to a detailed examination of the evidence and documents and was provided with 25 large ring-binders of what are said to be relevant matters. This Court does not exist for the re-ventilation of a dissatisfied litigant's complaint, but to correct errors of law. On the aspect of the copyright claim, I can find no error of law. The Judge applied the correct legal principles and saw and heard the witnesses. Although as I have said earlier, I may not necessarily have reached a similar conclusion on all the aspects of similarities, I am in no doubt that not only has derivation been proved, but also Dr. Lam has copied a substantial part of Dr. Koo's questionnaire. He has utilised someone else's skill and labour and unfairly produced it as though it was his own.

86. The section on passive smoking is important and was one of the original ideas worked on by Dr. Koo. Whilst copyright protects not ideas but expressions of thought, Dr. Lam has, in my judgment, copied Dr. Koo's expression of thought on passive smoking, although he has used different words. On this aspect alone the plaintiff has proved that the defendant has infringed a substantial part of her questionnaire.

87. Despite the able and attractive submissions of Mr. John Griffiths Q.C., I am in no doubt that this appeal on issues of fact has no merit at all and far from the Judge ignoring or misunderstanding the evidence, he has carefully sifted the relevant considerations from the mass of fact and opinion presented to him. It must be remembered that the Judge found the defendant to be an untruthful witness and it is clear from the transcript that the defendant had by circling or underscoring questions C12 and C13 in a copy of Dr. Koo's questionnaire which he had obtained, also considered the section on passive smoking as important. Although, in his own questionnaire, he has sought to disguise his copying of this section, in my judgment, as I have said earlier, it is strikingly similar.

88. Mr. Griffiths has submitted that the Judge's findings were solely directed to proving that one questionnaire was derived from the other. With respect, I do not agree. The Judge recognised that the facts had to support both causal connection and striking similarity.

89. Mr. Fung, Q.C., in very lengthy submissions, has taken the Court through large quantities of the transcript and documents to demonstrate that the Judge came to correct factual conclusions. Whilst I appreciate the importance of this appeal to his client, in my judgment, this Court should not give the appearance of re-trying the action. It is for the appellant to demonstrate that the Judge, who had the privilege of hearing and seeing the witnesses, was plainly wrong.

90. Despite the extensive skeleton arguments presented to this Court and the protracted submissions over 3 weeks, in my judgment, on the issue of breach of copyright, the appellant has palpably failed to persuade me that the Judge did not utilise the advantage he had of hearing the evidence in coming to the conclusion that Dr. Lam infringed Dr. Koo and Professor Ho's copyright in their literary work.

91. Even on a perusal of some of the transcript, it is clear to me that the Judge took an active part in this trial and was well aware of all the subleties and details of the evidence and the issues at which the evidence was aimed.

92. I turn then to the second ground of appeal relating to the Judge's finding that Dr. Lam "surreptitiously obtained" the questionnaire and misused the confidential information it contained.

93. Again the findings of fact are clearly supported by the evidence. The real complaint by Dr. Lam is that not only has the Judge branded him a liar, but also a thief.

94. I would gratefully adopt what my Lord, Penlington J.A. has concluded on the confidence claim.

95. I would only add this - Dr. Lam has only himself to blame for facing this claim. He had, on any view, no right to have in his possession Dr. Koo's questionnaire let alone utilize it. The Judge held that his possession of the document was not only unauthorised but he had "surreptitiously" obtained it. This may, in my judgment, connote an element of dishonesty. In Exchange Telegraph Company Limited v Gregory & Co. [1896] 1 QB 147, a similar finding was made by Lord Esher M.R., in a copyright claim, against a man who's surreptitious taking of confidential information was described as a "dishonest and contemptible thing". The facts of that case were, however, quite different. In the present instance the word "surreptitiously" means no more than that Dr. Lam had made unauthorised use of the questionnaire.

96. In his judgment, the Judge hoped that the bitterness between the parties would cease and "I dare to hope that they will by now have had their fill of litigation and are longing to get back to their useful work..." Unfortunately those wise and well-meaning words fell on stony ground and Dr. Lam launched a 3-week appeal as he was dissatisfied with the findings made against him.

97. I agree that this appeal should be dismissed.

Penlington, J.A.:

98. For the reasons given this appeal is dismissed.

99. Bokhary J. awarded the respondents their costs of the trial to be taxed, if not agreed, on a common-fund basis. He did so because he regarded the case as being unusual in the way the defence was conducted, in particular the manner in which it was denied that the appellant had made any use of the respondents' questionnaire until the 40th day of the trial when he admitted he had done so.

100. Those considerations do not seem to apply here as clearly there were very arguable grounds put forward for the appellant but the respondents may seek to persuade us to the contrary. There will be an order nisi that the respondents have the costs of the appeal to be taxed, if not agreed, on a party and party basis.