Products Liability in the Pharmaceutical Industry
at Common Law
Harvey Teff *
Introduction
as in the Commonwealth generally –
Discussion of the shortcomings of the common law action for
negligence in England –
has
been overwhelmingly concerned with injuries arising out of employ-
ment and road accidents. Products liability, in the sense of a manu-
facturer’s strict liability to the consumer for defective goods, has
received scant treatment.’ The citadel of privity may have been
taken by storm in America;2 the Englishman’s castle seemingly
survives intact.
However, whether the narrow conceptualism which denies the
consumer a right of action against the manufacturer in the absence
of negligence or contract will hold sway for much longer is seriously
in doubt. Law reformers in England are at last beginning to focus
attention on the implications of a products liability approach for
these matters. The Law Commission is currently examining the
subject 8 and a Royal Commission established in 1972 to investigate
the whole field of civil liability and compensation for personal injury
includes in its terms of reference “death or personal injury (includ-
ing ante-natal injury) suffered … through the manufacture, supply
or use of goods or services”. 4
Undoubtedly one major precipitating influence has been the
thalidomide disaster, which, because of its catastrophic dimensions
and much-publicised protracted negotiations, has generated discus-
sion of key issues bearing on products liability. A private member’s
Bill was introduced in Parliament in 1972 5 specifically to impose
e M.A., LL.M., Ph.D., Lecturer in Law, University of Durham.
‘See Legh-Jones, Products Liability: Consumer Protection in America, (1969)
27 Camb. LJ. 54; Jolowicz, The Protection of the Consumer and Purchaser of
Goods under English Law, (1969) 32 M.L.R. 1; Pasley, The Protection of the
Purchaser and Consumer under the Law of the U.S.A., (1969) 32 M.L.R. 241.
2 Prosser, The Assault upon the Citadel, (1960) 69 Yale LJ. 1099 and Prosser,
The Fall of the Citadel, (1966) 50 Minn.L.Rev. 791.
8 See the Law Commission: Seventh Annual Report (1971-72), 1-.M.S.O., 14,
para. 60.
4 The Pearson Commission. See H.C. Deb., Vol. 84, col. 1119-20, Dec. 19, 1972.
Dangerous Drugs and Disabled Children Bill, first reading, Nov. 29, 1972.
19741
THE PHARMACEUTICAL INDUSTRY
strict liability on the manufacturer of an unsafe drug, expressed to be
retroactive and to cover ante-natal injuries. Because of reservations
voiced about possible anomalies in piecemeal legislation responding
to a particular situation and in view of the subsequent course of the
negotiations, the Bill was withdrawn. It will, however, be suggested
that in several respects the thalidomide tragedy provides an un-
usually compelling demonstration of the potential benefits of apply-
ing a products liability test to the pharmaceutical industry.
Important underlying reasons for a reappraisal of the basis of
liability in this area are the steady erosion of fault liability elsewhere
in the common law world, reflecting mounting dissatisfaction with
the negligence action in general, and a parallel shift towards greater
consumer protection. The unwonted eagerness with which English
appellate courts are now articulating policies hitherto masked by the
all-purpose criterion of reasonable foreseeability has exposed the
artificiality and inherent contradictions of the negligence action,
while encouraging a more’ realistic attitude towards the role of
insurance.6
At the same time, “consumer law” is emerging as an area of
study in its own right, gradually being freed from that total identi-
fication with the world-of commercial contracts which has sustained
an unreal formal framework. The statutory recognition of the “con-
sumer” in the Supply of Goods (Implied Terms) Act, 1973,7 which
makes exclusion clauses in “consumer sales” void, and in the Fair
Trading, Act, 1973,8 which will enable the Director-General of Fair
6 See, e.g., Symmons, The Duty of Care in Negligence: Recently Expressed
Policy Elements, (1971) 34 M.L.R. 394, 528; Morris v. Ford Motor Co. Ltd.,
[1973] 2 All E.R. 1084, 1088; [1973] 2 W.L.R. 843, 846 per Lord Denning,M.R.
The House of Lords rejected this approach in Morgans v. Launchbury, [1973]
A.C. 127; [1972] 2 All E.R. 606 (H.L.).
7 S.4(7) provides that:
In this section “consumer sale” means a sale of goods (otherthan a sale
by auction or by competitive tender) by a seller in the course of a business
where the goods –
(a) are of a type ordinarily bought for private use or consumption; and
(b) are sold to a person who does not buy or hold himself out as buying
them in the course of a business.
8 S.137(2) states that:
… “consumer”… means any person who is either –
(a) A person to whom goods are or are sought to be supplied (whether
by way of sale or otherwise) in the course of a business carried on
by the person supplying or seeking to supply them, or
(b) a person for whom services are or are sought to be supplied in the
course of a business carried on by the person supplying or seeking
to supply them, and who does not receive or seek to receive the
goods or services in the course of a business carried on by him….
McGILL LAW JOURNAL
[Vol. 20
Trading to take injunctive action against manufacturers who consis-
tently produce shoddy goods, exemplify the first stages of a program
designed to transform this area of the law.
This recognition of an imbalance in bargaining power, coupled
with rising consumer expectations, reinforces the protection provided
by the Misrepresentation Act, 1967 and the Trade Descriptions Act,
1968, which went some way towards diminishing the purchaser’s
vulnerability in the face of sophisticated marketing techniques. In
addition to these legislative developments, the judicial creation of
liability for negligent mis-statement in Hedley Byrne v. Heller,9
with its implicit revival of non-contractual warranty, has provided
an as yet untapped source for strengthening the position of the con-
sumer. Our purpose is to examine the application of these trends to
products liability, with* particular reference to prescription drugs.”
Products Liability in America
The development of products liability in America has been fully
documented. 1 Though its details need not detain us, the routes which
have been followed and the emerging rationales contain invaluable
lessons for any common law country embarking on a similar venture.
We are fortunate in being able to draw on their experience in three
different lines of approach: raising the standard of care imposed on
the manufacturer in negligence; providing an action in breach of
warranty; and finally developing a strict liability doctrine in tort.
The judicial development of negligence in America, as in other
common law jurisdictions, has gone some way towards meeting
those critics who advocate a more comprehensive products liability
system. It is often claimed that the combination of higher standards
of care imposed on manufacturers and the doctrine of res ipsa
loquitur obviate the supposed need for reform. But this assertion is
surely unduly complacent, partly because the precise effect of res
ipsa loquitur remains unclear, but also because the litigation or
bargaining process may be profoundly influenced by the need to
prove negligence. Even if the onus of proof is shifted to the defend-
0 [1964] A.C. 465; [1963] 2 All E.R. 575. See also Stevens, Hedley Byrne v.
Heller: Judicial Creativity and Doctrinal Possibility, (1964) 27 M.L.R. 121, 160:
“[Hedley Byrne] may herald a whole new area of protection for the consumer.”
10Also known as ethical drugs, i.e., drugs not advertised directly to the
public.
11 See articles cited in f.n.1 and f.n.2, and also Kessler, Products Liability,
(1967) 76 Yale L.J. 887. The American literature is voluminous. In relation to
drugs, see especially Rheingold, Products Liability –
The Ethical Drug
Manufacturer’s Liability, (1964) 18 Rutgers L.Rev. 947.
19741
THE PHARMACEUTICAL INDUSTRY
ant, considerations of cost and delay will dissuade many claimants
from pursuing an action. Nor, in any event, does it seem particularly
healthy for the law to perpetuate as the basis of liability a system out
of touch with reasonable consumer demands and the economics of
modern marketing methods.
Breach of warranty has been a more fruitful source of develop-
ment. There is a wealth of case law, reinforced by section 2-318 of
the Uniform Commercial Code and similar legislative provisions,
imposing liability on manufacturers for breach of express or implied
warranty in the absence of privity and in the face of purported
disclaimers.
Yet however beneficial this approach may have been in particular
cases, because it was rooted in contract, inconsistencies and fictitious
elements emerged. If it entails classifying the supplier as a fictional
agent of the manufacturer or consumer, requiring the consumer to
give notice and indulging in over-elaborate argument on the precise
ambit of exclusion clauses, it is not easy to see the merit of asserting
even an attenuated contractual nexus between manufacturer and
consumer. This is especially the case when there are strong historical
grounds for the belief that the breach of warranty concept was tor-
tious in origin.’2 To adopt the same model would be to invite the
same consequences, with predictably more unfortunate results in
England, where traditionally there has been a marked opposition to
loosening the hold of privity.
It is certainly unsatisfactory to stress contractual elements in the
context of prescription drugs (at least where a Sale of Goods Act
model is relied on) since under the National Health Service there is
no technical “sale” to the patient, even where he pays a nominal sum
for the prescription.” The same problem is posed by free hospital
treatment, free drug samples and public vaccination programs. In
these situations recourse must be had to the device of relying on the
initial sale of the drug.14
It was due to such theoretical weaknesses that a strict liability
in tort theory eventually emerged. In three cases in 1913-14r, concern-
450 (H.L.).
12 See Stevens, f.n.9, supra, 162.
13 Pfizer Corp. v. Minister of Health, [1965] A.C. 512;
14 Cf. the American cases on blood transfusions, where liability was originally
denied because there was no sale, e.g., Perlmutter v. Beth David Hospital, 123
N.E. 2d 792 (1954). But see now Cunningham v. MacNeal Memorial Hospital,
47 Ill. 2d 443; 266 N.E. 2d 897 (1970).
[1965] 1 All E.R.
15 Mazetti v. Armour & Co., 75 Wash. 622; 135 P. 633 (1913); Parks v. G.C.
Yost Pie Co., 93 Kan. 334; 144 P. 202 (1914); Jackson Coca Cola Bottling Co. v.
Chapman, 106 Miss. 864, 869; 64 S. 791 (1914).
McGILL LAW JOURNAL
[Vol. 20
ing impure and adulterated food and drink, manufacturers were held
liable on broad grounds of public policy. The famous Henningsen
decision 10 in 1960, ostensibly based on breach of warranty, effectively
relied on changes in marketing techniques to justify strict liability
for all products, and was soon followed by a clear statement of strict
tort theory in Greenman v. Yuba Power Products Inc.1 7 A few years
later, in 1965, the Second Restatement of Torts 18 made far-reaching
provision for strict liability in respect of “unreasonably dangerous”
products. By 1971, approximately 30 States subscribed to strict
liability for manufacturers in general. 19
A tort framework has several advantages. It minimizes uncertain-
ty on whether a court will adopt a liberal or restrictive attitude
towards non-contracting claimants, and disposes of the requirement
of sale. It also makes it easier to ignore exclusion clauses, on the
basis that they are contractual terms which should protect the
defendant only in the context of contractual liability. 0 Perhaps
above all, it acts as a reminder that the principles governing liability
for personal injury in a sphere where the parties are not at arms
length should not be unthinkingly identified with those which
regulate commercial loss, particularly at the very time when, as
indicated above, consumer law is freeing itself from such identifi-
cation.2
The Current Position in English Law
How far does existing English law fall short of a strict liability
system?22 The consumer will not normally be in a position to sue
the manufacturer in contract, and third party proceedings are
acknowledged to be inefficient and costly, if not ruled out alto-
gether because of the insolvency of an intermediary or lack of
jurisdiction. A collateral contract has occasionally been invoked,
10 Henningsen v. Bloomfield Motors Inc., 32 NJ. 358; 161 A. 2d 69 (1960).
17 59 C. 2d 57; 377 P. 2d 897; 27 Cal. Rep. 697.
1s Section 402A.
10 Noel, Strict Products Liability Compared with No-Fault Automobile Acci-
dent Reparations, (1971) 38 Tenn.L.Rev. 297-98, n.4.
20 See, e.g., Vandermark v. Ford Motor Co., 61 C. 2d 256; 37 Cal.Rep. 896
(1964); and Legh-Jones, f.n.1 supra, 73.
21 Compare the Report of the Ontario Law Reform Commission on Consumer
Warranties and Guarantees in the Sale of Goods (1972), where an implied
warranty approach is advocated which would lead to “a substantial narrowing
of the gap between Ontario and American products liability law”. Ziegel,
(1973) 22 I.C.L.Q. 363, 364.
22 See Atiyah, The Sale of Goods 4th ed. (1971), chap. 13.
1974]
THE PHARMACEUTICAL INDUSTRY
most notably in Wells v. Buckland Sand and Silica Co.,-2 but this
has never been envisaged as encompassing widespread manufacturer
liability. The Wells case itself rested on an express warranty given
to the purchaser by the manufacturer: “It clearly emerges as being
quite fortuitous that … the defendants did not themselves sell
direct to the plaintiff. ’24 Moreover, collateral contracts and third
party proceedings are confined to plaintiffs who have bought the
product in question, thus retaining the disadvantages of contractual
solutions as outlined above.
If the contractual routes are unsatisfactory, is it nonetheless
possible to maintain that the existing law of negligence is adequate
to cater to the consumer’s legitimate demands? The argument runs
as follows. 25 In a negligence action you must prove three things:
that the product was defective; that the defect was present when
the product left the manufacturer’s premises; and that its defective
nature was due to the negligence of the manufacturer. The first two
requirements would have to be proven even under a products
liability system; the third –
is, it is said, largely
satisfied nowadays by the readiness of the courts to apply the
doctrine of res ipsa loquitur in manufacturer/consumer disputes.
negligence –
Unfortunately, however, the fault concept is still so deeply en-
trenched that even in cases where it appears on the facts to have
been abandoned, as in Grant v. Australian Knitting Mills, Ltd.26
and Mason v. Williams & Williams, Ltd.,2
7 the decisions have been
explained in accordance with orthodox negligence theory, and the
applicability of res ipsa loquitur has been denied. Further, the effect
of that doctrine in the English case law is far from settled.28 Often
23 [1965] 2 Q.B. 170; [1964] 1 All E.R. 41.
24 Ibid., at 177 and 43 respectively.
25 See Prosser, The Assault upon the Citadel, supra, 1114-1119.
26 [1936] A.C. 85; [1935] All E.R. 209.
21 [1955] 1 All E.R. 808; [1955] 1 W.L.R. 549.
28See Street, The Law of Torts 5th ed. (1972), 138.
The effect of res ipsa loquitur in Canadian law is not fully settled, but the
weight of opinion leads to the conclusion that once its requirements are met,
the plaintiff has succeeded, if only temporarily, in satisfying the onus of
proof. The defendant then will have to respond or lose his case: Temple v.
Terrace Transfer Ltd. (1966), 57 D.L.R. (2d) 631 (B.C. C.A.). However, it does
not seem that the defendant has the “onus of disproof”. Rather, the prevailing
view is that he may satisfy the onus now resting on him with a mere ex-
planation, consisient with the facts, of how the damage could have occurred
without his negligence: Temple v. Terrace Transfer Ltd.
There are cases which have shifted the burden of proof more forcefully
to the defendant. Thus in Zeppa v. Coca-Cola, [1955] O.R. 855 (Ont. CA.) it
McGILL LAW JOURNAL
[Vol. 20
the courts maintain that it is not a distinct rule of law shifting the
legal onus of proof onto the defendant, but merely permits an
inference of negligence. 29
However, in two recent House of Lords decisions, Henderson
v. Henry E. Jenkins & Sons and Colvilles, Ltd. v. Devine, the onus
was in effect shifted.30 As long as any doubt remains concerning
the full force of the doctrine, the most that can be said is that the
courts may interpret it in a manner which places a heavy burden on
the defendant. But even if they were to do this consistently, it would
not remove a most serious impediment to an acceptable solution
of any technically complex case in which negligence was in issue,
namely, the ability of a powerful defendant to compel settlement on
terms unfavourable to a plaintiff who cannot take the risks or
endure the delay which any investigation of that issue might involve.
As Tobin put it:
Our courts have not yet faced the difficult issues that are raised for
instance by prescription drugs. When they do, it is submitted that it will
be preferable to approach the question of the extent of manufacturer
liability in terms of the social and economic reasons for holding product
makers liable rather than in terms of “fault” alone.3′
As long as a negligence framework is retained, many judges will
inevitably approach consumer claims against manufacturers with
a notion of individual fault uppermost in their minds.
The Choice of Rationales
Apart from specific criticisms of the negligence action, perhaps
the three grounds most commonly cited in favour of a products
liability approach are deterrence, the moral responsibility of the
manufacturer and rational risk allocation. The deterrent argument
is in essence the popular notion that the stricter the liability to
was held that res ipsa loquitur had the effect of placing a burden on the
defendant to actively disprove negligence. From the point of view of the con-
sumer, this is an encouraging note, since it would clearly aid the consumer-
plaintiff in actions against manufacturers. However, this case represents an
exceptional departure from the more established rule in Temple v. Terrace
Transfer Ltd.
20 Lloyde v. West Midlands Gas Board, [1971] 2 All E.R. 1240; [1971] 1 W.L.R.
749.
30 Henderson v. Henry E. Jenkins & Sons, [1970] A.C. 282; [1969] 3 All E.R.
756; Colvilles, Ltd. v. Devine, [1969] 2 All E.R. 53; [1969] 1 W.L.R. 475. See
Atiyah, Res Ipsa Loquitur in England and Australia, (1972) 35 M.L.R. 337,
340-344.
3 l Tobin, Products Liability: A United States/Commonwealth Comparative
Survey, (1969) 3 N.Z.L.R. 377, 401-402.
19741
THE PHARMACEUTICAL INDUSTRY
which he is subject, the more care the manufacturer will take to
see that his production methods are safe. Moral responsibility
primarily connotes his obligation to stand behind products which
through advertising are aimed at consumers as a class. It is also
implicit to some extent in the “hazardous enterprise” theory: “these
liabilities are the price which must be paid to society for the per-
mission of a hazardous activity”.32 On yet another level, it is the
corollary of making profits under a free enterprise system.
The other major approach centres on a more clinical search for
rational risk distribution. Nice questions of allocation of responsi-
bility are overriden. Accident losses are deemed an inevitable price
of doing business and the manufacturer is normally able to calculate
the risks more efficiently than the potential victim. Liability insur-
ance and costing are brought out into the open, instead of being
concealed within an apparently fault-based system.
However, the very process of labelling and separating these
various justifications is quite artificial. It understates the essentially
pragmatic way in which many cases are decided and may convey
an impression of mutually exclusive or necessarily competing ex-
planations. The outcome of particular cases may owe something to
the emphasis placed on particular rationales, but one should not
lose sight of the extent to which they reinforce one another. Public
policy, moral obligation and risk distribution were all explicitly put
forward as justifications in Escola v. Coca Cola Bottling Co.3 4 In
so far as products liability arguably provides a stimulus to safety
in manufacture, however minimal, it can only add force to this
argument.
Products Liability and the Pharmaceutical Industry
Assuming that legal deterrence in the pharmaceutical industry
is felt by manufacturers, a change from a negligence to a products
liability system would hardly be more inhibiting to them. Both
systems are in practice underpinned by insurance. The first line of
attack would surely be more stringent governmental requirements
prior to marketing. Such controls are most notably evident in the
United States’ Federal Drug Authority regulations. 5 Britain, which
32 Ehrenzweig, Negligence Without Fault (1951), 54.
33 No attempt has been made here to illustrate the range and complexity
of risk distribution theory. For a recent analysis see, e.g., Calabresi & Hirschoff,
Towards a Test for Strict Liability in Torts, (1972) 81 Yale L.J. 1055, 1071.
3424 C. 2d 453, 461-462; 150 P. 2d 436, 440-441 (1944) per TraynorJ.
35The American regulations are arguably too inflexible. Under the Food
responsible for the ban on
Additives Amendments 1958 (as amended) –
McGILL LAW JOURNAL
[Vol. 20
in the past relied on a voluntary system, has now introduced a
measure of compulsion 0 Safety procedures in most countries have
been tightened up considerably since the thalidomide disaster 7
More generally, however, there is a familiar and more plausible
explanation of safety levels than the fear of civil actions, or indeed
of any legal intervention. Just as one drives safely in order to stay
alive, so a company’s degree of care in production methods is largely
dictated by its wish to survive and create goodwill, and as such is
governed by the ordinary controls implicit in running an effective
business 8
A more convincing case for products liability can be made out
on grounds of moral obligation and risk distribution. But here also
their detailed application raises serious difficulties. In the case of
prescription drugs, the consumer/patient is to a large extent relying
on the middleman/doctor, for a product not directly advertised
to the public. This necessarily complicates the issue if one is em-
phasizing responsibility for marketing the product. A comparable
problem is posed when a drug manufacturer abroad grants a
licence for the distribution of a compound containing its product,
yet retains a measure of control over the sales promotion techniques
employed. This, in fact, was how thalidomide was distributed in
several countries.
If, on the other hand, one adopts a resource allocation model
and takes as a starting point the proposition that for maximum
efficiency the pharmaceutical industry should pay its own accident
losses, several key issues are left in the air. What products are
involved? If we confine strict liability to drugs, how do we define
“drug”? If we extend liability to pharmaceutical products in general,
is it feasible to work out a rational domestic form of risk distribu-
tion when dealing with huge, multi-national, highly diversified
concerns, which can vary their production and marketing methods
cyclamates in 1969 –
any substance found to be carcinogenic in animals is
prohibited. Hence saccharin, used for nearly 100 years without the slightest
evidence that it is cancer-producing in man, is in danger of being taken off
the market, regardless of the potentially more harmful consequences of in-
creased sugar consumption.
3o Medicines Act, (1968) 16 & 17 Eliz. II, c.67.
37 In Canada, increased concern is reflected in the Federal Government’s
QUAD (Quality Assessment of Drugs) Program. See Comments on Drug
Quality and the Quad Program, Hon. Marc Lalonde, Minister of National
Health and Welfare, 13 March 1973 (News Release).
3 8 This does not rule out fear of the possible impact of civil liability on
trading reputation as one motivating factor. See Linden, Tort Law as Ombuds-
man, (1973) 51 Can. Bar Rev. 155.
1974]
THE PHARMACEUTICAL INDUSTRY
so as to take advantage of national tax laws which are most favour-
able to them?
Even those American States which have gone “all the way” with
products liability have encountered intractable problems with drugs.
It is true that the historical progression from impure and adulter-
ated food and drink to animal food to products for external bodily
use would suggest that drugs and medicines have strong claims if an
order of priority were to be drawn up for strict liability treatment.
But this does not allow for the special market structure of the
industry, the unique method of distribution in the case of ethical
drugs and the complex nature of the products and their effects. Even
the simplest drug transaction is apt to involve problems of causa-
tion which far outweigh the difficulty of deciding whether the
plaintiff lost his finger because of a faulty machine.
What bearing does the economic structure of the pharmaceutical
industry have on the degree of liability to which it should be sub-
ject? The plea is occasionally heard that this is an industry working
uniquely for the benefit of mankind, as selfless dedicated healers,
and as such must not be discouraged from carrying out its creative
role by harsher laws. But antibiotics are not discovered every day
and the bread and butter, routine work involves slight changes to
established formulae, the end product being just sufficiently dissim-
ilar to obtain a patent –
the “close copy” products born of
molecular roulette. Indeed, to the extent that consumers are being
wooed with promises of “instant relief”, the process is more in-
sidiously commercial than many other industries.
Intense commercial pressures operate at every stage of market-
ing a drug. Patent life under English law lasts for sixteen years, 9
but in practice profitable patent life is no longer than ten or twelve
years, since much time may be absorbed in developing the product
pharmacological and toxicity testing, clinical trials and registra-
–
tion –
and the final years are undermined by competitive develop-
ments which force the price down shortly before the patent expires.
On the other hand, the goodwill created by strong brand promotion
lives on, especially if the product has enjoyed a near monopoly of
the market.
Producing a genuinely new drug, as distinct from a close copy
product, requires a vast initial outlay, recently estimated at 3 to
4 million in the United Kingdom and at 6 to 9 million in the
United States.40 Only one in approximately five thousand compounds
39Patents and Drugs Act, (1949) 12 & 13 Geo. VI, c.62, s.22(3).
4oSee Beckett, “The cost of safety in medicines” in Teeling-Smith (ed.),
The Pharmaceutical Industry and Society (1972), 19.
McGILL LAW JOURNAL
[Vol. 20
tested will ultimately find its way on to the market. There is then an
immense pressure to produce one’s product first and a psychological
as well as commercial drive to market it rather than abandon costly
research, reinforced by a resistance to removing it from the market
until conclusive evidence of harm has been provided. It is therefore
hardly surprising that the promotional element looms large in the
manufacture of drugs.41
The profit ratio of pharmaceutical products is very high in
comparison with industry in general and with other parts of the
chemical industry in particular.42 To what extent such profits are
necessary to recoup past research and development expenditure and
to provide adequately for future innovation is the central issue in the
current litigation between the British Government and Hoffmann-
, Certainly there seems to be little evidence of a direct
La Roche.4
relationship between successful innovation and specific levels of
research and development spending. Britain’s efficiency record has
recently been calculated as two and one half times that of the United
States, proportionate to the amount invested in pharmaceutical
research.44
Indeed beyond a certain level research expenditure could un-
justifiably prejudice existing patents for the sake of speculative
future gains. We simply do not have an objective basis for deciding
proper pricing and research funding criteria for pharmaceuticals,45
and in practice we lack the financial data to make an adequate
evaluation when dealing with diversified, multi-national companies.
The Hoffmann-La Roche litigation provides a classic illustra-
tion of these difficulties. The concern is the largest pharmaceutical
dealer in the world, with a controlling centre in Switzerland and
subsidiaries in many countries, including Canada, England and
the United States. The company has a well-earned reputation for
41 Total expenditure on sales promotion of National Health Service prescrip-
tion drugs in the United Kingdom for 1965 was calculated as 13.9 per cent
of total sales: Report of the Committee of Enquiry into the Relationship of
the Pharmaceutical Industry with the National Health Service, 1963-65 (the
“Sainsbury Report”), H.M.S.O. (1967).
42 In 1968-69 it was 8 to 10 per cent above the “total industries” average in
the Financial Times’ “Trend of Industrial Profits” table, and 12 per cent above
the corresponding figure for the chemical industry as a whole.
-43 See Department of Trade and Industry v. F. Hoffmann-La Roche and Co.
A.G. and Ors., The Times, 14 July 1973; F. Hoffmann-La Roche and Co. A.G. and
Ors. v. Secretary of State for Trade and Industry, The Times, 31 July 1973.
44Innovative activity in the pharmaceutical industry, a Report by the
Pharmaceuticals Working Party of the Chemicals EDC, H.M.S.O., 1973, 4, 25.
4 5 Vaizey, in preface to The Pharmaceutical Industry and Society, see fm.
40, supra.
1974]
THE PHARMACEUTICAL INDUSTRY
innovation, quality of research and safety of its products. Its dis-
covery some ten years ago of new sedatives – Librium and Valium
– meant that doctors no longer felt compelled to prescribe batbi-
turates which have potentially more harmful side-effects.
When the Government was concerned at the price the company
was charging for these tranquillizers, an investigation by the
revealed profits -on sales of up to 60
Monopolies Commission”
per cent and rates of return on capital of over 70 per cent. Thus
a highly-regarded company with a virtual monopoly of the market?
and protected by patents 4 was subject to little or no direct con-
sumer pressure on prices. At the same time it was able to maximize
profits by adjusting its operations to take account of differing
national taxation systems.
The Government ordered 49 the company to reduce the price
of Valium to “not more than 25 per cent of the selling prices in
1970” (not more than 40 per cent in the case of Librium) and to
repay alleged excessive profits. The company, which has appealed
against the Order, 0 maintains that the Government’s policy will
stifle initiative at a time when research and development are
becoming prohibitively expensive. Whatever its merits in this case,
it is submitted that this argument is unconvincing as a general
proposition. On-going research is of the essence of a pharmaceutical
company’s activities. Even when no specific breakthrough results,
goodwill which helps the company to negotiate licences is main-
tained, and the continued production of close copy drugs is facili-
tated.
The overall picture of the pharmaceutical industry is therefore
one of high profitability. The increasing costs of marketing and of
research and development inhibit the entry of new companies into
the .market, and one might reasonably anticipate more and more
concentration of resources among huge international concerns.
Most companies would seem to be in a position to adjust relatively
painlessly to a strict liability system (as the American ones already
have) and insure for all but the most catastrophic of losses. Nor
46 Chlordiazepoxide and Diazepam, H.M.S.O., 1973.
47Approx. 99 per cent, ibid., 57, para. 194.
48Its hold on the market -was reinforced by the free distribution of Librium
and -Valium to hospitals and the armed forces, making their continued pre-
scription highly probable.
– 49The Regulation of Prices (Tranquilising Drugs) Order, 12 April 1973,
renewed on 18 May 1973 and 21 June 1973.
5OThe Times, 31 July 1973. See f.n.43, supra.
McGILL LAW JOURNAL
[Vol. 20
would the marginal increase in cost to the consumer, who in Eng-
land spends on the average about one penny a day on National
Health Service medicines, prove unduly burdensome.
Admittedly there is a problem when one considers new drugs
with a potential for causing catastrophic loss. The actuarial process
operates by reference to the past. It may be prohibitively expensive
to insure against some drug risks if in fact they are insurable at
all. This difficulty and its implications will be considered when
assessing the outcome of the thalidomide negotiations. 51
The special method of distribution of prescription drugs is
another distinctive feature of the industry. One might in the ab-
stract expect a drug manufacturer to be subject to strict legal
control because of the intrinsic dangers of his products and the
above average vulnerability of his clientele.5
Is this argument
substantially affected by the role of the doctor? He is not a mere
conduit in the chain of supply, but has an independent, profession-
al duty to evaluate. Even if the exigencies of general practice tend
to make prescribing a rather mechanical process, there is still
much greater reliance on the “seller” than is the case with most
goods. Nor should a doctor be allowed to exculpate himself entirely
by claiming that he prescribed on the basis of a company’s mis-
leading promotional material, for that is to abdicate his respon-
sibility. In fact most doctors probably place more reliance on
discussions with professional colleagues than on any other source
as a guide to prescribing, alihough promotional material and
medical sales representatives must exercise some influence.
Since it is the doctor who primarily determines which pills the
patient consumes, the marketing of prescription drugs also appears
to lack the important feature, from a moral obligation standpoint,
of direct advertising to the public. But one cannot divorce a
company’s reputation as an ethical drug manufacturer from its
general reputation, if doctors are to continue prescribing its prod-
ucts. The general motive of fostering reliance to increase sales
remains.
Perhaps the most intractable problems which any system of
liability for drug-induced injury must face relate to causation.
Typically, drugs are taken by people who are already ill and often
unusually susceptible to further ailments. Unlike many other prod-
ucts, they may cause injury in unpredictable ways, depending on
the user’s individual constitution. He may be allergic to a particu-
lar drug. On the other hand, what appears to be an allergy may in
SAt pp. 121-22.
52 This also, of course, implies a high standard of care in negligence.
19741
THE PHARMACEUTICAL INDUSTRY
fact be a toxic reaction. Presumably a balance must be struck
between a defence for the manufacturer based on allergy and the
conflicting tort principle that you must take your victim as you
find him.
Further difficulties stem from the ways in which drugs are
taken. The synergistic effects of certain combinations (for example,
barbiturates and alcohol, anti-histamines and cheese) may prove
fatal. Harmful self-administration may take several forms, such
as overdosage, unduly prolonged use, failure to inform one’s doctor
of a previous medical condition or taking drugs stored in medicine
cupboards long after there has been publicity about their po-
tential danger.
Even when properly administered, many drugs will have pos-
sibly harmful side-effects but be deemed medically acceptable be-
cause on balance their use is beneficial. This peculiarity raises a
problem concerning the definition of “defect”, the hallmark of
products liability. The American Restatement (Second)53 conclud-
ed that drugs which are typically unavoidably unsafe, yet socially
indispensable, should not be considered “defective”. But as Mr
Justice Traynor has pointed out, we should arguably be more
ready to impose liability where the product’s norm is danger, this
being classic enterprise theory, and in particular where a reason-
ably safe substitute exists5 4
Finally we might note in passing the anomalous position of the
patient who has become addicted to a drug, or in the case of ciga-
rettes who has contracted lung cancer. In these situations it is
precisely as a consequence of using the manufacturer’s product
over a period of time that the injured party is often prevented from
protecting himself. The addiction model is illuminating in that it
illustrates, in an extreme form, the vulnerability of drug con-
sumers as a class. One possible inference is that defences such as
disclaimer, volenti and contributory negligence should be con-
strued very strictly against drug manufacturers.
It is tempting to conclude that as products liability is no better
equipped to cope with the above kind of causal problems than
negligence, a wider criterion still is needed on the lines of a state
insurance scheme of the New Zealand type.55 This alternative
53 American Restatement (Second), Torts, para. 402A, comment k.
54The Ways and Meanings of Defective Products and Strict Liability, (1965)
32 Tenn.L.Rev. 363. Cf. Ehrenzweig, f.n.32, supra.
55 Accident Compensation Act, 1972. See also .The Report of the Royal Com-
mission of Inquiry: Compensation for Personal Injury in New Zealand, N.Z.
Government Printer, 1967 (the “Woodhouse Report”).
McGILL LAW JOURNAL
[Vol. 20
cannot be dealt with in detail here, but it should be borne in mind
when considering the history of the thalidomide dispute in England
as elsewhere.
The Thalidomide Dispute
Thalidomide is a sedative invented by the West German com-
pany, Chemie Grunenthal. In England it was manufactured and
distributed by Distillers Company (Biochemicals) Ltd. (DCBL) in
1958 under the trade name of Distaval, and under licence from
Chemie Grunenthal. In 1961 DCBL withdrew the drug from the
market following evidence that if taken in the early stages of
pregnancy it could cause serious injury to the embryo.
Sixty-two actions were brought against the company within the
limitation period (i.e., within three years of the birth of the
children). But before they were heard, the plaintiffs withdrew all
allegations of negligence and settled on the basis that the defend-
ants would pay 40 per cent of the damages awarded to all plaintiffs
who were not statute-barred. In 1968 Hinchcliffe, J. approved
settlement on these terms in principle, but since an agreement was
not
reached on quantum, two cases were tried in 196950 in the
nature of representative actions. A child described as in the middle
category of injured was awarded 32,000 (and his mother 7,500)
and a child in the most serious category received 52,000. In both
cases 60 per cent was deducted.
The judgment exemplified some of the features of the existing
common law action which are most open to criticism. The judge
regretted that so much time had been taken up in dealing with
questions of actuarial evidence.5 7 The view that inflation and in-
terest should be deemed to cancel one another out was adopted.5,
A global sum was awarded, based on experience and calculated to
be fair to the plaintiffs and to the defendants.
Meanwhile many more cases were gradually coming to light,
and by 1971 it appeared that there were more than 400 claims not
covered by the 1968 settlement. In November, 1971 DCBL proposed
to set up a charitable trust fund of 31/ million for these children,
56S. v. Distillers Co. (Biochemicals) Ltd., [1969] 3 All E.R. 1412; [1970] 1
W.L.R. 114.
r7 [1969] 3 All E.R. 1412, 1421. On the basis of actuarial evidence submitted,
the awards (subject to the 60 per cent deduction would have been approxi-
mately 78,500 for the middle category and 135,000 for the most serious
category. See Prevett, Actuarial Assessment of Damages: The Thalidomide
Case –
II, (1972) 35 M.L.R. 257, 263.
Is Cf. Mallett v. McMonagle, [1970] A.C. 166, 177; [1969] 2 All E.R. 178, 190.
19741
THE PHARMACEUTICAL INDUSTRY
provided that all the parents concerned accepted. Five families
refused the offer, and an application to have them replaced by
the Official Solicitor as next friend was dismissed by the Court
of Appeal.5 9 The negotiations were resumed.
In September, 1972 the Sunday Times published the first of a
proposed series of articles on the subject. Two months later an
injunction was granted restraining publication of one of these
articles, which purported to show that DCBL had not exercised
due care to see that Distaval was safe before it was put on the
market. The basis of the decision was that the article formed
part of a deliberate campaign intented to influence the course of
the litigation and amounted to contempt.60 Towards the end of
November, coverage of thalidomide in the press was at its height
and the campaign to provide a better settlement for the children
had reached national proportions. On 28 November, the day before
Parliament was to debate the issue and hear the First Reading of
the Dangerous Drugs and Disabled Children Bill, the company
increased its offer from 31/4 to 5 million.
In January, 1973 it seemed that the Privy Council was about to
be called upon to decide whether a child has a right of action in
tort for ante-natal injury, as the Supreme Court of Victoria had
held in Watt v. Rama,”‘ but the defendant did not appeal. In the
same month the Law Commission expressed a preliminary view in
favour of such a right.6 2 Also, the company substantially increased
its offer to 20 million.
In the following month the Court of Appeal discharged the
injunction against the Sunday Times, mainly on the ground that
the litigation was dormant. However, the House of Lords subse-
quently allowed an appeal from this decision,6 stressing that nego-
tiations for a settlement were being actively pursued and that the
proposed article contained material expressly aimed at establish-
ing that the company was negligent.
59Re Taylor’s Application, [1972] 2 Q.B. 369; [1972] 2 All E.R. 873.
60 Att.-Gen. v. Times Newspapers Ltd., [1972] 3 All E.R. 1136; [1972] 3 W.L.R.
855. But cf. Att.-Gen. V. London Weekend Television, [1972] 3 All E.R. 1146;
[1973] 1 W.L.R. 202, where a television program shown one week later was
not deemed to create “a serious risk that the course of justice would be
interfered with”.
61 [1972] V.R. 353.
62 Law Commission Working Paper No. 47: Injuries to Unborn Children.
Att.-Gen. v. Times Newspapers Ltd., [1973] 1 All E.R. 815; [1973] 2 W.L.R.
452.
298 (H.L.).
4 Att.-Gen. v. Times Newspapers Ltd., [1973] 3 All E.R. 54; [1973] 3 W.L.R.
: McGILL LAW JOURNAL
[Vol. 20
In April, 1973 the company put forward an offer which consti-
tuted the basis of the final settlement, 5 and which has been accept-
ed by 95 per cent of the parents –
accounting for 433 children.
Each child who qualifies,00 but did not share in the 1968 settlement,
is to recover 40 per cent of the value of his claim assessed as of
1970, at a rate comparable to the 1968 cases, plus 33 per cent to
take account of inflation and interest. All the children, including
those in the 1968 settlement and any who later qualify, will share
in a charitable trust fund of 14 million, to be paid in seven yearly
instalments of 2 million, protected against inflation by up to
10 per cent compound interest. In addition, each pair of parents
is to receive 5,000. On average this settlement currently’
repre-
sents approximately 54,000 for each family –
as compared with
8,000 on the basis of the 3 1 million offer made in November,
1971.
In one sense, the end of a long, painful and tortuous path has
been reached, resulting in awards which will be “virtually equiva-
lent to a settlement on a full liability basis”.6 8 But there is little
cause for complacency when one reflects on the inordinate delay
(without prejudice to responsibility for it in the instant case)
which must have contributed to the family breakdowns and
disruptions experienced by many thalidomide families; 9 the risk
that a few families took in holding out against the 31/ million
offer; the apparent necessity of an intensive press campaign; and
the fact that it may still be several years before individual settle-
ments are finalized. Not least disconcerting from a legal point of
view is the absence of any guidelines in the event of a comparable
future disaster.
What then are the legal issues involved in drug-induced catas-
trophes of this kind?70 If pleading negligence, a claimant would
have to show that a cause of action exists in respect of ante-natal
G5 Allen and Ors. V. Distillers Co. (Biochemicals) Ltd., The Times, 31 July 1973.
0 Le., broadly speaking, anyone who is accepted as a thalidomide child by
DCBL or satisfies a judge that his injuries were caused by any pharmaceutical
preparation manufactured by DCBL in the United Kingdom, containing thali-
domide.
07A residual problem concerns over 100 disputed claims which, if they were
all ultimately acknowledged as valid, would reduce the average award by about
9,000.-
08 See f.n.65, supra.
60 H.C. Deb. Vol. 847 No. 22 col. 470.
70 See Bennett, The Liability of the Manufacturers of Thalidomide to the
Affected Children, (1965) 39 A.LJ. 256.
1974].
THE PHARMACEUTICAL INDUSTRY
injury and that the manufacturer (or distributor, as the case may
be) 71 was in fact negligent in marketing the drug.
Little needs to be said here about ante-natal injury.72 There is
no authoritative English case law on liability for it in tort, and
the Irish case, Walker v. Great Northern Ry. Co. of Ireland,”
denying a right of action, is hardly persuasive, influenced as it was
by unscientific fears about the difficulty of proving causation. 4 A
growing body of Commonwealth authority7 5 and the generaltrend
in the United States 76 are now in favour of a remedy. The cumula-
tive effect of professional and academic support,7 public sympathy
and the Pearson Commission’s terms of reference 78 suggests that
such a right will soon be definitively established by legislation.
It is on the fundamental issue of negligence that the thalido-
mide litigation has proved most revealing. As negotiations and
proceedings in several countries have shown 7 9 any claimant faces
71 The theoretically possible but much weaker case against the prescribing
doctor will not be considered.
72 A useful summary is to be found in the Law Commission Working Paper
No. 47, see f.n.62, supra.
73 (1891), 28 L.R. Ir. 69; 8 Digest (Repl.) 618.
74 (1891), 28 L.R. Ir. 69, 82.
75 E.g., Montreal Tramways v. LdveilM, [1933] 4 D.L.R. 337 (Que.); Duval v.
Seguin (1972), 26 D.L.R. (3d) 418; Watt v. Rama, [1972] V.R. 353; Pinchin v.
Santam Insurance Co. Ltd. (1963)
(2) S.A. 254 (W).
76Twenty-one states .now allow recovery and only two deny it. See also
White v. Yup, 458 P. 2d 617, 620-621 (1969): Law Commission Working Paper
No. 47, supra.
77 The Law Society has added its support to the Law Commission’s provi-
sional recommendation: The Times, 11 June 1973. See also Winfield, The
Unborn Child, (1942) 4 U. of T. LJ. 278.
78 At p. 102.
79 The German criminal proceedings (including preliminary investigations)
lasted for eight and a half years before being suspended, as their continuation
was deemed contrary to the public interest. For an interesting account of
these proceedings and of thalidomide generally see Sjostrom & Nilsson,
Thalidomide and the Power of the Drug Companies (1972).
In Canada, many -settlements were reached in 1968, allegedly worth from
100,000-150,000 (Sunday Times, 22 April 1973). Parents who did settle were
obliged not to disclose the amounts. But other suits alleging negligence have
been filed quite recently in United States’ jurisdiction on behalf of children
from Quebec, where limitation provisions prevented them from suing.
In Australia, after some seven years of negotiations, the company has made
offers averaging 51,000 to 13 claimants, plus 5,400 for each set of parents:
Sunday Times, 15 July 1973.
In Japan, actions begun in 1965 are still continuing. The company marketing
the drug has denied that thalidomide causes deformity: Sunday Times, 6 May
1973.
McGILL LAW JOURNAL
[Vol. 20
the prospect of a prohibitively cosdy and unendurably lengthy
action. One could not construct a more vivid model to confound
those critics of products liability who maintain that it already
exists in all but name. For in the thalidomide cases, unlike pre-
natal injury claims based on nervous shock, causation in the vast
majority of cases is beyond dispute, 0 so precisely are the conse-
quences calculable. Proving negligence, on the other hand, would be
a long drawn-out, expensive struggle. Even if res ipsa loquitur is
interpreted as placing the burden of proof on the defendant, he
could mount a highly technical case,81 asserting the adequacy of
his safety precautions and tests in the light of the then current
procedures, and invoking as evidence of reasonableness the degree
of governmental sanction which attached to the drug in question. 2
Clearly a products liability system would greatly enhance the
plaintiff’s prospects of success, and at the very least make litiga-
tion a practical proposition, with the further likelihood that a
realistic settlement might be achieved without undue delay. A
possible objection to liability along the lines envisaged by the
Second Restatement, para. 402A comment k (“new or experimental
drugs as to which because of lack of time and opportunity for
sufficient medical experience, there can be no assurance of safety”)
could be rebutted where (as in the case of thalidomide) accept-
able, safe substitute drugs are in existence.
The instinctive popular reaction to the thalidomide tragedy
was essentially based on the notion of moral responsibility. It was
reiterated in the Commons Debate:
… the powerful and wealthy Distillers Company, with assets of 64
million, has had no compunction in fighting these children for no less
than 10 years … a display of moral irresponsibility which has seldom if
ever been surpassed.83
As we have seen, the moral obligation approach derives much of
its force from the view that a manufacturer ought to stand behind
his products. Thus it was suggested in the Debate that “[t]he
Company aggressively marketed thalidomide”,8 and much atten-
80 Apart from an equivocal reaction from Japan’s Health Ministry, every
drug-controlling authority has recognized this in respect of thalidomide.
81 As happened in the German criminal proceedings; see f.n.79, supra.
82 As was pointed out in the Parliamentary Debate, thalidomide had been
classified under “New remedies of proved value”, by the Standing Joint Com-
mittee on the Classification of Proprietary Medicines. H.C. Deb. Vol. 847 No.
22, col. 435.
82Ibid., col. 434.
84 Ibid., col. 433.
19741
THE PHARMACEUTICAL INDUSTRY
tion has focused on the advertisement current in October, 1961
that “Distaval can be given with complete safety to pregnant
women”.
From the perspective of risk, one might simply argue that the
individual should not bear losses of this order because he is not
in a position to calculate the risk as efficiently as the manufac-
turer. If it be said that the manufacturer may not be able to
calculate the risk either, a compromise solution would be possible
using Ehrenzweig’s “typicality test”8 5 (imposing liability where the
injury is broadly typical of the kinds of hazard to which the
enterprise gives rise). No doubt there are difficulties concerning
the precise level of generality covered by such a test, but the basic
idea is clear. Damage to the foetus resulting from the use of a
particular drug may be wholly unexpected, but should still import
liability, being a side-effect of a specifically medical character. Such
a test can be justified in that it requires separate enterprises to
bear their own accident losses, and emphasizes the desirability
of calculating costs as far as possible prior to engaging in an
enterprise, so that one may then insure accordingly.86
However, while the general financial state of the pharmaceutical
industry is probably healthy enough for it to absorb and insure
against normal losses, thalidomide has reminded us that the scale
of mass-production of drugs is such that a single error can affect
vast numbers of people throughout the world and that there is no
guarantee that a given company will be able to insure against, or
even obtain coverage for, such an eventuality. The argument for at
least partial public responsibility in such situations can be sup-
ported on the ground that a government seal of approval should
import a measure of financial responsibility.87
Products liability is not the only alternative to negligence. If
one regards such injuries mainly as a price of technological advance
and stresses the benefits of medicine to the public in general, 8 a
85Ehrenzweig, Negligence Without Fault (1951).
86 See James, The Untoward Effects of Cigarettes and Drugs: Some Reflec-
tions on Enterprise Liability, (1966) Cal.L.Rev. 1550.
87 Cf. The Pharmaceutical Journal, (1973) Vol. 211, No. 5727, 84: “the govern-
ment should accept at least joint responsibility with a manufacturer, whose
product has met standards of safety set by the government, but from the
use of which any injury or loss has arisen”.
88 “[I]n the case of vaccinations, compulsory or recommended, the state
under German law, is liable for damages”: Bundesseuchengesetz, July 18 1961,
(1961) 1 BGBL 1012. Quoted, Caemmerer, chap. 4 of Aspects of Comparative
Commercial Law, Ziegel & Foster (ed.)(1969).
McGILL LAW JOURNAL
[Vol.’20
social security model of risk allocation based on general taxation
might seem most appropriate. Certainly where there are losses of
massive proportions, some governmental backing would be justifi-
able to meet the needs of the injured and to prevent reputable
firms from going bankrupt.
The introduction of a partial or
total state compensation
scheme would obviously raise large issues of social and political
philosophy, but its desirability would also partly depend on how
effectively the industry could function under a products liability
system. In 1964 Rheingold said that “‘t]his’ subject is badly in
need of a socio-economic study to determine factually what the
impact of strict liability would be upon the drug industry”.89 The
current investigation into the affairs of Hoffmann – La Roche is one
aspect of what should be an on-going process of analysis, and
itself demonstrates the difficulty of formulating objective criteria
to resolve the problem.
Conclusion
Whether in respect of normal or catastrophic loss the negligence
action for injury caused by drugs has proved to be inefficient,
primarily for procedural reasons. There is therefore a cogent case
for introducing a system of products liability. While in itself this
may not sound very startling, it must be remembered that the
concept is one which, when first introduced into a legal system,
is likely to encounter strong opposition not only, from the industry,
but also from those who would see in it an unacceptable abdication
of individual responsibility 0
Further, available financial data suggests that the industry is
capable of bearing the increased burden which this would entail in
the ordinary course of events. To provide for catastrophic losses,
it might be possible to make compulsory insurance up to a given
amount a prerequisite of the right to market a drug, with the
government underwriting any excess.
This discussion has been largely concerned with prescription
drugs. The conclusions, therefore, are open to the objection that
they would lead to piecemeal reform producing yet further anom-
8) (1964)
18 Rutgers L.Rev. 947, 1015, n.359. See the “Sainsbury Report”
(f.n.41, supra) for a detailed analysis of the industry in England during the
period 1963-65.
00 For a recent exposition of this view, see Stoijar, Accidents, Costs and Legal
Responsibility, (1973) 36 M.L.R. 233.
19741
THE PHARMACEUTICAL INDUSTRY
aliesY1 However, always assuming that products Riabiiity is the
preferred alternative to negligence, this form of criticism is not
simply too perfectionist, but fundamentally misconceived. For it
should be evident that there can be no all-purpose formula for
products liability. The particular form
takes in any given
industry will have to be worked out on the basis of as thorough
an analysis as is practicable of the peculiarities of its market
structure.
it
91 Cf. Richard Crossman (in an article on disabled children): “By the acci-
dent that a wealthy company could be held responsible (sic) for their mis-
fortune, they (the thalidomide children) will become a tiny plutocratic elite
while 8,500 children (many of them with disabilities just as grave as theirs)
are condemned to lifelong incarceration in obsolete long-stay hospitals.” The
Times, 18 July 1973.
