The Implications of Science for the Law

[Vol. 18

The Implications of Science for the Law

Peter Brett *

It has become a truism –

almost a clich6, indeed –

to remark
that we are living in an age of science. Nevertheless I am constrained
to begin this essay by uttering that truism. For I wish to argue
that the law –
our profession, our art, our discipline, call it what
you will –
has grown out of touch with the scientific outlook
that dominates our age, and that it must change its attitude so
as to harmonise with that outlook if it wants to remain a living
and respected force in contemporary society.

At the outset of the Gifford Lectures which he delivered in
1959-60, the eminent physicist Carl-Friedrich von Weizs~icker 1 dis-
cussed the interrelationship of science and religion. He pointed out
that religious agnosticism is, most probably, the dominant attitude
of the western mind in our time. The years which have passed
since he spoke have confirmed rather than diminished the force
of this observation. Of course we have recently seen the emergence
of cults ranging from witchcraft and satanism to the “Jesus freaks”,
but these are really no more than excrescences. They do little or
nothing to satisfy on a large scale the psychological hunger of
human beings for something to believe in and to guide their lives by.
That hunger, von Weizsiicker thought, was being filled in our time
by science, or, as he preferred to put it, by “scientism” –
faith
in science.

It is interesting that the cults which I have just mentioned
are at present showing signs of increased popularity and that at
the same time we can discern in the western world signs of a
reaction against the dominance of science. This may have occurred
because, to quote von Weizsdcker again, “the guiding factor in
faith is not belief but trust”. The object of faith needs to be trust-
worthy –
to demonstrate that the reliance placed on its powers
is not misplaced. It is precisely because organised or institutional
religion lost this quality of trustworthiness over a long period
of time that it has ceased to be a dominating force; such calamities
as the Lisbon earthquake of 1755 dealt it blows from whose effects

* Professor of Jurisprudence, University of Melbourne.
1 C. F. von Weizsdcker, The Relevance of Science, (1964). All references are

to chapter 1.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

it never really recovered, and the ability of science to make good
its promises drove the lesson home.

In recent years science itself appears to have been treading
the same path of distrust and rejection. In the late nineteenth and
early twentieth centuries it rode the crest of a wave of optimism.
The prevailing mood was one of belief that every problem could
be solved by science, which was leading us rapidly towards a paradise
that we were just on the point of entering. The claims of scientists
which provided the underpinning of this mood, though understand-
able enough if considered in their historical and social context,
were too large. It is one thing to have the ability to find solutions
to a problem, and quite another to choose the wise solution from
among the possibilities. Questions of value enter into the act of
choice, and one of the features of nineteenth and early twentieth
century science was its affirmation that it was not concerned with
choice and evaluation –
that its sole task was to provide expla-
nations and techniques and that it was for others to decide how
those explanations and techniques should be used.

Here again we are witnessing a change of mood or of heart;
scientists today are examining their responsibilities to the society
in which they live. Meanwhile, the technologically-based countries
are struggling to deal with the problems of pollution, population
increase, and so on which the uncritical use of technology has
produced, and not surprisingly there is a temporary disenchantment
with science. I say a “temporary” disenchantment because scientism,
like any other faith, will not disappear until something better
appears to displace it. And I can see no reason to suppose that
such “faiths” as occultism, satanism, or the “Jesus revolution” will
prove adequate replacements. After all, they were in the past
weighed in the balance and found wanting (for the purposes I have
been discussing). Moreover it is an observable fact that the societies
which have not yet reaped the benefits of modern science and
technology are most anxious to do so, despite the accompanying
perils; it is the technologically-advanced societies that are showing
disenchantment, and even then only with the evils that uncontrolled
technology brings in its train. Technology is, in truth, something
like a remedial drug in the armoury of weapons against the diseases
of life as it might be lived by men in the “state of nature” envisaged
by Hobbes. We would prefer to have a drug without serious side
effects, but we would nevertheless rather have a drug with side
effects than no drug at all. Hence the solution to our present
problems seems to lie in the hope of improving and controlling
technology –

not in repudiating and abandoning it.

McGILL LAW JOURNAL

[Vol. IS

What follows is written with these beliefs in mind. Despite
the temporary reaction in some quarters against the dominance
of science, I believe that during any future period with which the
readers of this essay are likely to be concerned the methods of
science will be a major influence in guiding our course. I further
believe that an institution or system of social control that disregards
or turns its back on the method of science is likely to wither away.
For it is the method of science that provides the basis for our
trust in its powers, and we are not likely to continue to endure
an institution in which we can no longer place our trust.

II

At this point I should discuss a probable objection. “Non-
sense”, the reader may say. “How can it be claimed that the law
is out of touch with the scientific spirit and discoveries of the age?
A glance at the various law reviews circulating throughout the
common-law world is enough to refute any such thesis. The reviews
are packed chock-a-block with articles discussing the application
and use of scientific discoveries and techniques for legal purposes.”
I readily concede the apparent force of this reply. Indeed, if I
were willing to devote a great amount of time to no useful purpose,
I could at this point add a footnote which would extend over many
pages or tens of pages, giving citations to this material. Except
for the occasional, very rare instance, however, I do not regard
the extra-legal material discussed in these articles, and sometimes
in the court decisions, as dealing with science. It would be more
accurate to say that what is invoked and discussed is pseudo-science.
To ascertain the scientific status (or otherwise) of a particular
theory we can fruitfully make use of the tests propounded by Sir
Karl Popper, the noted philosopher of science. Not only is he a
recognised authority in his own field, but his views are accepted
by eminent scientists 2 who have concerned themselves with the
matter as being an accurate description of how scientists behave.
Popper’s criterion of the scientific status of a theory is its fal-
sifiability, or refutability, or testability. In a lecture delivered in
1953, 3 he explained that he had originally formulated this criterion

2 Such as Sir John Eccles and Sir Peter Medawar (both Nobel prize-winners

for medical research), and Michael Polanyi (a physical chemist).

3 Originally entitled “Philosophy of Science: a Personal Report”, republished
as chapter 1 of his Conjectures and Refutations, (1963), esp. pp. 36-38. See
also: Popper, The Logic of Scientific Discovery, (1959).

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

some 35 years earlier as a summary of considerations which had
led him to the following conclusions:

1. It is easy to obtain confirmations, or verifications, for nearly

every theory –

if we look for confirmations.

2. Confirmations should count only if they are the result of
risky predictions; that is to say, if, unenlightened by the theory
in question, we should have expected an event which was incom-
an event which would have refuted the
patible with the theory –
theory.

3. Every “good” scientific theory is a prohibition: it forbids
certain things to happen. The more a theory forbids, the better
it is.

4. A theory which is not refutable by any conceivable event is
non-scientific. Irrefutability is not a virtue of a theory (as people
often think) but a vice.

5. Every genuine test of a theory is an attempt to falsify it,
or to refute it. Testability is falsifiability; but there are degrees
of testability: some theories are more testable, more exposed to
refutation, than others; they take, as it were, greater risks.

6. Confirming evidence should not count except when it is the
result of a genuine test of the theory; and this means that it can
be presented as a serious but unsuccessful attempt to falsify the
theory (Popper now terms confirming evidence which satisfies
this criterion “corroborating evidence”).

7. Some genuinely testable theories, when found to be false,
are still upheld by their admirers –
for example by introducing
ad hoc some auxiliary assumption, or by re-interpreting the theory
ad hoc in such a way that it escapes refutation. Such a procedure
is always possible, but it rescues the theory from refutation only
at the price of destroying, or at least lowering, its scientific status.
(Popper describes a rescuing operation of this kind as a “con-
ventionalist twist” or a “conventionalist stratagem”).

Popper explains that at the time he was engaged in formulating
this test he applied it to several theories: specifically, Einstein’s
theory of gravitation, astrology, the Marxist theory of history,
the psycho-analysis of Freud, and the individual psychology of Adler.
The latter three theories came under his scrutiny because they were
being widely discussed among students at the time; this perhaps
explains why the analytic psychology of Jung (which should surely
accompany the theories of Freud and Adler) was not included.
Of the five theories considered, Einstein’s passed the test easily
(Eddington’s eclipse observations in 1919 had brought important

McGILL LAW JOURNAL

[Vol. 18

confirmation of it), but the remaining four failed. The predictions
of astrology were so vague that they could scarcely fail, and were
thus irrefutable. The Marxist theory of history was testable, and
in some respects its predictions had proved false; but by re-
interpreting the theory and the evidence (the conventionalist strata-
gem) the supporters of the theory rescued it from refutation.

The theories of Freud and Adler were of a different kind, for
they were, and still are, completely non-testable. It is impossible
to conceive of human behaviour that would refute them. They
thus cannot lay claim to scientific status. Exactly the same point
has been made in the last few years by Sir Peter Medawar, as
regards psycho-analysis and the much more recent existentialist
school of thought in psychiatry.4

To the list of theories just discussed we could add some others
which are currently enjoying a great vogue as all-embracing ex-
planations. The behaviourist theory of B. F. Skinner has temporarily
superseded psycho-analysis (or so we are sometimes told), and in
the field of sociology “alienation” and “urbanisation” are making
strong claims to be the root of all evil. Since these theories exhibit
exactly the same defects as those listed by Popper –
vagueness
of concepts, irrefutability, re-interpretation ad hoc –
we are
justified in rejecting their pretensions to be scientific as opposed
to mythological.

This last remark needs a qualification as regards behaviourism.
It originated as a methodology in the study of psychology, and
there can be no complaint as to its scientific status so long as it
remains a methodology. As Medawar neatly puts it:

Behaviourism… taught psychology with brutal emphasis that ‘the dog
is whining’ and ‘the dog is sad’ are statements of altogether different
empirical standing, and heaven help psychology if it ever again overlooks
the distinction.5
The advocates of behaviourism have, however, gone far beyond
treating it as a methodology; they have erected it into a theory, and
in doing so have eradicated the distinction that Medawar makes.
Starting from the fact that ‘the child is weeping’ is a very different
type of statement from ‘the child is sad’, they have moved to
arguing that behaviour such as that described by
‘the child is
weeping’ is all that exists, and that states of mind, feelings, purposes,

4 Medawar, Science and Literature, Encounter (Sanuary, 1969), 15, at p. 22.
See also: Medawar, The Art of the Soluble, (Pelican, 1967), and Medawar,
Induction and Intuition in Scientific Thought, (Methuen, 1969).

G Medawar (1967), ibid., at p. 100.

No. 21

IMPLICATIONS OF SCIENCE FOR THE LAW

and so on (as in ‘the child is sad’) are fictions which mankind
no longer needs. Behaviour is all; it can be fully described and
explained in terms of stimulus and response, and it is completely
determined by the environment, i.e. everything outside the skin
of the human being that (I could not, in this context, properly
is behaving. Thus Skinner was reported recently as
say ‘who’)
saying:

Consider a woman who has a baby. It cost her a lot of pain and trouble
to have it. But she didn’t design that baby; it was all settled at the
moment of conception what the baby was going to be like. The same
thing is true when a man writes a book, invents things, manages a
business. He didn’t initiate anything. It’s all the effect of past history
on him. That’s the truth, and we have to get used to it.6
I quote this passage because it illustrates some of the difficulties
that behaviourists find themselves in. The speaker has not managed
to eradicate references to allegedly mythical persons, thoughts,
and feelings, and it is highly improbable that such a task could
be performed. Moreover, it is hard to see what force should be
given to the last sentence of the quotation. Skinner’s utterances
are, on his theory, merely a part of my environment, and as such
they have no claim to preference over the contrary utterances of
an anti-behaviourist.

There is no need for me to set out the reasons which make
behaviourism as a theory not only impracticable but indeed meaning-
less; the only point I wish to make is that its irrefutability (in so
far as it can be said to have any real meaning) robs it of any
pretension to scientific status. And I would add, in passing, that
if all other objections to it could by any chance be overcome,
there would be some formidable scientific problems for it to meet.
It is easy enough for a behaviourist to assert that no man initiates
that his brain acts simply as a machine for processing
anything –
stimuli into responses. But there is powerful evidence from neuro-
physiologists that the brain acts in quite a different way.7

indeed, for society as a whole –

In one respect, however, behaviourism may well pose a problem
for lawyers –
that could be more
troublesome than any problem posed (up to the present time at
any rate) by the other theories I have mentioned. For it has given
rise to a dubious practice called “aversion therapy”. The nature
of this is explained by Professor B. F. Singer, an experimental

6 Time Magazine, vol. 98, (September 20, 1971), at pp. 49 et seq. See also:
Skinner, Science and Human Behaviour, (1953) and Skinner, Beyond Freedom
and Dignity, (1971).

7 Sir John Eccles, Facing Reality, (1970).

McGILL LAW JOURNAL

[Vol. 18

psychologist, in an article intended to inform lawyers of recent
“advances” and accordingly published recently in a law review.”
It merits a detailed comment.

For some years, experimental psychologists have been investi-
gating the behaviour of rats, with a view to ascertaining the ef-
fectiveness of what they term “punishment”. The design of the
experiments was pioneered by Skinner. A good deal of information
has been accumulated. The rats are isolated in a situation where
they have the opportunity of pressing a lever; if they press it they
get an electric shock and so they quickly give up pressing it.
If time is allowed to elapse between pressing the bar and receiving
the shock, the rat takes longer to give up the practice. Again, the
severer the shock on pressing the bar, the more quickly the rat
gives up. It has even been found that an animal subjected to
treatment of this kind which is sufficiently painful will starve to
death rather than repeat the behaviour that produces the pain.

These may be important scientific discoveries of the twentieth
century. I am inclined to think, however, that they do not provide
any information that was not known centuries ago to circus animal
trainers. At the present time, the trainers are prevented by laws
about cruelty to animals from acquiring further experiences that
would complement the experimental observations. The laws probably
do not apply in scientific laboratories or protect rats. Nevertheless
it may be that the experimenters have slighly uneasy consciences,
since they use the term “punishment” to describe the shocking of
the rat. It seems a curious use of the wordY

There might be some difficulty in applying the information
derived in this way to human beings, arising from the fact that the
human brain is so vastly different in its structural complexity from
that of a rat or other lower animal as to merit the comment that
it is different in kind rather than in degree.10 (I do not think, inci-
dentally, that Professor Singer mentions this fact.) Leaping over
this difficulty with gay abandon, some psychologists and psy-
chiatrists have devised the technique known as “aversion therapy”,
which works in the same way as the rat experiments. Thus if we
wish to stop a man from sucking his thumb, we can wire him up
to a machine which will give him an electric shock every time he
puts his thumb in his mouth. If he co-operates with the therapist in

8 B. F. Singer, Psychological Studies of Punishment, (1970), 58 Calif. L. Rev.

405 et seq.

9 M. Ginsberg, On Justice and Society, (London, 1965), chapter 10.
10 Eccles, op. cit., n. 7, chapter II.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

a series of sessions during which he puts his thumb in his mouth
repeatedly and each time gets a shock, he soon learns not to suck
his thumb. It is not, at present, known whether he will retain the
lesson for the rest of his life, and it may well happen that he takes
to sucking his little finger instead. Nevertheless, it can be claimed
that he has been cured, at least temporarily, of his thumb-sucking;
and if need be we can repeat the process and cure him of his little-
finger sucking, his middle-finger sucking, his straw sucking, and
so on –

if we and he do not get bored.

Professor Singer shows how “aversion therapy” can be applied
to “cure” criminal behaviour. He discusses a case in which a habitual
woman shoplifter sought help by its use. The initial efforts failed,
but:

The therapist then repeatedly showed the patient a treatment film in
which a woman entered a co-operating store while a number of people
in the store overtly watched. When the woman then shoplifted several
items on a counter, the film showed expressions of horror and disgust
on the faces of those watching. At the moment the woman in the film
shoplifted the items, the therapist shocked the patient. Hospital personnel
(nuns) administrated the treatment. This treatment was successful. The
patient finally stopped shoplifting, and she reported uneasy feelings
of being watched whenever she entered a store. The therapist planned
to repeat the treatment every few months.11
I must confess to some difficulty in reconciling the final sentence
of this extract with the claim that the treatment was “successful”.
Perhaps the explanation is that the therapist did not have many
patients – which would not be surprising, even though he had
the help of nuns to show that what he was doing was ethically
acceptable. Another explanation could be deduced from Singer’s
remark, in the paragraph that follows this extract, that:

no one really knew how long the treatment would be effective. It might
have lasted only a few months or a lifetime. Since stores are such
common environmental objects, extinction might soon set in.12

[“Extinction” here refers to the effect of the treatment on the
shop-lifting, not to the patient.]

This seems to suggest that the criterion of “success” is some-
thing other than permanent suppression of shop-lifting tendencies
on the patient’s part. However, in a footnote 13 Singer reports that
is
“Bandura [the author of a book on behaviour modification]
optimistic about the potential durability of such treatments, given

11 Singer, op. cit., n. 8, at p. 432.
12 Ibid., at p. 433.
13 Id., footnote 106.

McGILL LAW JOURNAL

[Vol. 18

the proper procedures.” This no doubt rallies the reader’s hopes,
though Singer almost at once dashes them again by noting the
patient’s later uneasiness whenever she entered a store. In the
next sentence, however, he raises them again by remarking that
the uneasiness, though “certainly a personal handicap”, was not as
great a handicap as the patient’s “previous affliction”, and that
such a result is “part of the price of successful aversion therapy”. 14
At this point I add that it is an inevitable price. It arises in this
instance because “shoplifting” has been “treated” by methods which
ignore what a lawyer would call the mens rea of the shoplifter.
Of course, the behaviourist must ignore this, because mental states
do not enter into his universe of discourse. But it seems obvious
that they must enter into any meaningful description of criminal
conduct. One can test this by endeavouring to distinguish between
the activities of a woman who, while the shop assistant is elsewhere,
picks up an article on the counter and puts it in her shopping basket,
intending to carry it to the cashier’s desk and pay for it, and those
of another woman, who similarly picks up a similar article and
puts it in her shopping bag intending to steal it. The distinction
must of course be made without reference to the intentions of either
woman. I do not believe that this can be done, and I am confident
that a criminal code could not be written in terms which eschewed
any reference to mental states.

Singer is not, apparently, troubled by these problems, for he
follows his description of the shoplifting case by an enthusiastic
vision of the future:

One can treat almost any criminal behaviour in a similar manner.
Particularly when working with a prisoner serving a sentence of a year
or more, the behavior therapist would have time to discover the most
effective treatment, and having found it, he could make the conditioning
thorough. The program might require community co-operation. For
example, a large banking enterprise could establish a branch office for
the aversive conditioning and
robbers. Correction
personnel could administer most of the program. Given the time and
the resources, a behavior therapy program could make a bank robber
want to vomit every time he saw a bank, could make an armed robber
shudder every time he saw a gun. As experimenters have successfully
conditioned verbal and imaginal stimuli, such a program could also
induce these reactions whenever the convicted thief even thought or
talked about guns and banks. The program could include booster treat-
ments after prison whenever needed. Afterwards, the offender should
probably do his banking by mail.15

testing of bank

14 Id.
15 Id.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

The date when we could hope to implement such a programme
is not suggested, but I should imagine it will be some time after
1984. Ond I should warn that there may be some problems in
enlisting the aid of the correction personnel (perhaps nuns will
once again help out) and of the rest of the community, including
the bank officials. After all, they may have to endure a lot of
unpleasant sights and smells.

The reader may think that in the last-quoted extract Singer
is being humorous. In my estimate, however, both the immediate
context of the extract and the rest of the article give the lie to
any such explanation.

Up to this point, Singer has been talking about “aversion therapy”
as it has hitherto been practised, that is, as a form of “treatment”
carried out at the request and with the co-operation of the patient.
Almost at once, however, he mentions that “while a voluntary
aversion therapy program would probably be most effective, in-
voluntary treatment could also work. Such therapy could be part
of a regular sentence or an alternative to it.” 16

Here the wheel has finally come full circle. There is really nothing
new about the use of physical torture as a means of special de.
terrence. In the past it was far more used than imprisonment,
though the favoured methods were the lash, the pillory, the stocks,
and the ducking stool. Since it was difficult to exert close quanti-
tative control over the amount of pain inflicted –
a point which
troubled Bentham 17
it is understandable that contemporary
enthusiasts prefer to use electric shocks and induced nausea, thus
allowing us refinements of control unknown to our predecessors.
Just as modern technology has produced this benefit, so also modern
techniques of jargon permit us to disguise what we are doing by
such reassuring words and phrases as “patient”, “treatment”, “cure”,
and “involuntary aversion therapy”. I prefer the older methods as
at least more honest.

–

After all this, the reader will not be surprised to learn that
among the possibilities for improvement of our present unscientific
and ineffective methods of dealing with convicted criminals, Singer
recommends consideration of the following: requiring the offender
to make a public apology to his victim who would then cere-
moniously [sic] forgive him; requiring the offender to wear a
scarlet letter; and the revival of stocks and dunkings. He consoles

10 Ibid., at p. 434.
17 Bentham, Principles of Penal Law, Part II, Book II, chapter I, found in

Works of Jeremy Bentham, (Bowring ed., 1843), I, at pp. 414-415.

McGILL LAW JOURNAL

[Vol. 18

us with the remark that some of these treatments ( he does not
say which) “are in many respects more humane than prison” and
that they might be more effective (the latter point appears to
ignore the experience of the past),’ s

I have selected this article by Singer for critical discussion
because it exhibits most of the vices that the lawyer should be on
his guard against. The trivial and well-known is disguised in the
trappings of scientific language and set forth as if it were a new
discovery; the basic theory (behaviourism) is quite untestable but
advanced as if it were established scientific fact; and every possible
problem of jumping from the behaviour of rats in laboratories
to that of human beings in society is glossed over with the statement
that the “scientific” laws of behaviour can only be thus extrapolated
if all other things are equal (though it is surely obvious that when
the jump is made other things will not and cannot be equal). In-
consistencies in the available data that would put a genuine scientist
on his guard are relegated to footnotes, 9 and above all there is the
nauseating disclaimer of moral responsibility by means of frequent
reminders that the ethics of what might be done is a matter for
“society” and that the scientist is “no more qualified than any other
man” 20 to place a value judgment on particular applications of
his knowledge (one might conclude from this remark that he is
also no less qualified and no less gubject to moral duties). I find
this last point peculiarly distressing and repulsive. I have no use
for torturing people to get evidence, but I would find a mediaeval
treatise on the rack more honest and decent than, say, an experi-
mental psychologist’s paper on “the elicitation of information by
involuntary aversive stimuli”, especially if the latter were accom-
panied by disclaimers of moral responsibility.

Let me, however, make it quite clear that my criticisms are
not directed to all forms of experimental psychology. On the
contrary, I have no doubt that much experimental work in psycho-
logy is of great value, and that psychology could not hope to
progress as a science if it were not firmly based on experimental
work wherever possible. Our problem, as lawyers, is to distinguish

1s Singer, op. cit., n. 8, at p. 435.
19Thus Singer, ibid., footnote 112, at p. 434, refers to recent work demon-
strating the possibility of “conscious control over brain waves”. It does not
seem to occur to him that such conscious control is irreconcilable with the
view that human behaviour is completely determined and explainable solely
in terms of stimulus and response.

20Singer, ibid., at p. 410.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

between the genuinely scientific work in the field and the remainder,
to make use of the former, and to reject the latter.

I should like at this point to make a brief reference to another
species of pseudo-science which is showing signs of intruding into
the administration of justice. A trial is a form of historical enquiry
and thus its outcome is necessarily a matter of uncertainty. We
can never recapture the past, and so our results will always leave
us with the uncomfortable feeling that we may have been wrong.
Accordingly, there have been some recent efforts to lessen this
discomfort by the recommendations of techniques, based on math-
ematics, which will supposedly enable us to place more confidence
in our assessments. One well-known move of this kind has been the
introduction of “prediction tables” into the process of sentencing
convicted offenders 21 A more recent move is the effort to make
use of Bayes’s theorem in combining the inferences which can be
drawn from various items of evidence.22

I believe that many of these recommended mathematical or
statistical techniques are basically unsound. I do not wish to cast
doubt on the value of statistical tables when properly used. But
statistics are collections of data about groups of people or events
with common characteristics, and while they can properly be used
to make inferences about the future behaviour of similar groups,
they tell us nothing about the future behaviour of any individual
member of the group. Nevertheless, the belief that they can be
used to predict the outcome of individual instances persists, as
the wealthy proprietors of gambling casinos can attest. 3

The Bayes’s theorem type of error is of a different order. Once
again, I do not wish to cast doubt on the validity of the theorem,
which I gther has long been in use by statisticians. It can be said,
however, that the mathematical formulae used in calculating proba-
bilities are all based on arithmetic, and involve dealing with items
that can be counted. For instance, in dealing with the throws of
dice we start with the fact that the six faces of each die can be
counted one by one. We can also count the number of instances
in a series (albeit long or short) of throws. And so far as I can

21 For discussion see papers in E. Glueck and S. Glueck, Ventures in

Criminology, (Cambridge, Harv. U.P., 1964).

22L.H. Tribe, Trial by Mathematics: Precision and Ritual in the Legal
Process, (1971), 84 Harv. L. Rev. 1329. Professor Tribe advances excellent
reasons for rejecting the use of this technique as too dangerous in practice,
but I do not understand him to argue (as I do) that it is theoretically unsound.
23 H. C. Levinson, Chance, Luck and Statistics, (Dover ed., 1963) provides

a full, simple and non-mathematical discussion of these matters.

McGILL LAW JOURNAL

[Vol. 18

judge, the Rev. John Bayes in his theorem pre-supposed that the
prior but unknown probabilities which concerned him could be
computed (at least in theory) by counting instances.

The legal literature on this topic uses the Bayes theorem in a
different way. It combines what are termed “subjective proba-
bilities” with “objective” ones.24 The latter are ascertained (at least
in theory) by counting instances, but the former cannot even in
theory be ascertained in this way. They are no more than an assess-
ment, in numerical terms, of the strength at which the believer
assesses his belief. Thus, if his belief in an inference is very weak
he might assess the likelihood that the inferred event had occurred
as being 1: 1000, whereas if it were a much stronger belief he might
assess it as 1: 50. But in neither case would the figure be arrived
at by counting a series of instances, for in fact the event under
consideration would (if it had occurred at all) be an unique instance.
From this it will be seen that the numbers used in assessing
subjective probabilities are no more than symbols used to determine
a ranking order. It thus seems totally improper to use arithmetical
techniques in combining such numbers with numbers (used in
assessing objective probabilities) ascertained or ascertainable by
counting. The former are assigned quite arbitrarily and are easily
replaceable by letters or other synmbols; the latter are not.25

24 Tribe, op. cit., n. 22, at p. 1348 points out that the justification for doing
this is that subjective probabilities obey the same rules as objective proba.
bilities, “once a few entirely plausible postulates are accepted”. Whether these
postulates are entirely plausible or should be accepted is, however, a matter
of debate. Among them is a postulate of transitiveness, i.e. that if A believes
event X to be more probable than event Y, and event Y to be more probable
than event Z, he must therefore believe event X to be more probable than
event Z. No doubt A ought to think in this way, but it cannot be assumed
that he does. Human beings are often irrational.
25 Whitehead, in one of his last lectures, made it very clear that there is
a vast difference between the behaviour of numbers which are operated
arithmetically and the behaviour of what the numbers represent. In Immor-
tality, The Interpretation of Science, (1961), he says, at p. 266:

Let us take the simplest case; for example, the sentence, ‘One and one
make two’.
Obviously this sentence omits a necessary limitation. For one thing and
itself make one thing. So we ought to say, ‘One thing and another thing
make two things’. This must mean that the togetherness of one thing
with another thing issues in a group of two things.
At this stage all sorts of difficulties arise. There must be the proper
sort of things in the proper sort of togetherness.
The togetherness of a spark and gunpowder produces an explosion, which
is very unlike two things. Thus we should say ‘The proper sort of to-

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

Despite these criticisms, which have been made time and again
by others far more eminent and expertly qualified than I am, I
expect to see the pages of legal literature continue to contain
large amounts of pseudo-science and pseudo-mathematics. It is not
hard to see why this should be so. The material that I have been
discussing is comforting. It gives an appearance of introducing
certainty. or near-certainty into realms which have hitherto been
full of doubt. In doing so, it meets what seems to be a basic need
of human beings. 6

Moreover, some of this material provides further advantages. The
general theories of psycho-analysis, behaviourism, alienation, and
so forth have, as Popper has said,27 all the qualities of a religious
revelation. Wherever the convert looks, he can find instances of
their applicability and explanatory power. And those of them which
reject considerations of morality and responsibility as outworn
fictions not only provide ammunition for certain classes of lawyers
(such as criminal defenders) but also comfort every one of us
when we transgress (as, sooner or later, we all do).

Again, such explanations as “alienation” or “urbanisation”
comfort us, paradoxically, by offering no hope of a practicable
solution to certain evils that beset us, such as the burgeoning of
drug addiction. There are some practicable steps we could take
to lessen, if not to eradicate, many of these evils, but they are,
as a rule, unpalatable steps. It is preferable to avoid them and
talk, instead, about the “inevitable fruits of urbanisation” or what-
ever. We can thus do nothing at all and at the same time have
an excuse for our inaction. To say this is not to deny the existence
of problems arising from alienation or urbanisation. My criticism
is directed to the use of these concepts as all-embracing explanations
for every form of social evil.

Let us, however, return to the proper use of science in the
administration of justice. As I have said, it can and should play
a most valuable part. But there are dangers. I shall briefly mention
two that have attracted my notice, adding the warning that there
are probably several others.

The first and most obvious danger is that genuine scientific
material is often unfamiliar and hard to understand. It is therefore

getherness of one thing and another thing produces the sort of group
which we call two things’.
“proper sort of togetherness”.

My point is that subjective and objective probabilities do not have the

26 G. Yahoda, The Psychology of Superstition, (1969), esp. chapter VIII.
27 Popper, op. cit., n. 3, at pp. 34-35.

McGILL LAW JOURNAL

[Vol. 18

usually, perhaps always, desirable to have it critically scrutinised
by experts in the relevant field in order to avoid the possibility
of error. The danger of misinterpretation is especially acute where
reliance is placed on secondary sources 2 8 or where the material is
of the kind which is familiar and thus regarded as being of common
knowledge (examples are interpretations of statistical data and
the elementary uses of probability calculus). 0

The second danger is that scientific material is always pro-
visional and is constantly becoming out-of-date, so that yesterday’s
truth is today’s error. Unfortunately, however, in the law yesterday’s
belief, when embodied in a binding precedent, becomes authority for

28 Cf. Evatt, J. (dissenting) in Chester v. Waverley Corporation, (1939), 62
C.L.R. 1 (High Ct.) at pp. 14 et seq. At p. 47 the learned judge argues that
fright or nervous shock causes actual physical damage, and quotes in support
a note in 11 Can. B. Rev., at pp. 516-517. The writer of that note (and,
following him, Evatt, J.) appears to be confusing the two types of shock
described in medicine as “primary” and “secondary”: see K. Bowden, Forensic
Medicine, (2nd. ed., 1965), at pp. 250-252.

29 Commonwealth v. Malone, (1946), 354 Pa. 180, 47 A. 2d. 445, is the most
startling instance known to me. The Supreme Court of Pennsylvania there
upheld the conviction of the defendant (aged 17) for the second-degree murder
of his friend. The two youths, who had somehow obtained a five-chambered
.32 revolver and a cartridge, agreed to play “russian roulette” (described in
the report as “russian poker”). The defendant placed the cartridge in the
chamber to the right of the firing pin, placed the revolver against his friend’s
side, and pulled the trigger three times. On the third pull the weapon dis-
charged and caused a fatal wound. The report does not show whether the
cylinder was twirled between each pull, but does show that the defendant
testified that he had no intention of harming his friend and did not expect
the gun to go off.

The Court upheld the conviction with the remark that the fatal pull of
the trigger was done intentionally “in reckless and wanton disregard of the
consequences which were at least sixty per cent certain from his thrice
attempted discharge of the gun”. It seems clear that they were greatly im-
pressed by their “sixty per cent certainty”. But 60% is an impossible figure;
the only possibilities are 20%, 25%, 331/3%, 50% and 100%. We can discount
the last of these on the ground that although it was in fact correct it was
clearly not realised to be so by the defendant. Of the remaining four choices,
the most likely seem to be either 20% or 331/3%, though the meagre informa-
tion in the report does not aid one to decide the matter. The Court seem
to have been impressed by the fact that 60% is more than one-half, and it is
reasonable to suppose that they would have taken a different view if they
had assessed the probabilities correctly.

I may add that I have taught this case to a class of American graduate
students, some of whom had studied elementary probability problems in
their undergraduate courses. Not one of them noticed the Court’s error until
it was pointed out.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

today.3 Hence courts may, with the best of intentions, find them-
selves unwittingly -perpetuating what the world of science now
regards as error.

These dangers should not lead us to distrust scientific data and
evidence –
they should warn us merely to exercise caution and
to keep up-to-date in our use of scientific material. But while we
can and should make use in our work of the help that genuine
contemporary science can afford, an even more useful aid for us
is that of scientific method. The essential feature of this is that
the scientist adopts in his work a critical as opposed to a dogmatic
attitude.3′ It is this approach to the universe that accounts for his
visible success and gives rise to the trustworthiness of his work.
And if we wish to obtain a similar trustworthiness, we should
use his method.

I shall now give some examples of how this can be done.

III

There are four major features of our legal system which can
be properly described as fundamentally unscientific in spirit. They
are:

1. the adversary system of trial
2. the basic postulates of the rules of evidence
3. the portrait of human functioning which underlies many, in-

deed most, of our substantive rules of law and

4. the rule of stare decisis

I shall consider them in that order.

As Professor Sawer points out:
Court procedures are often classified as either inquisitorial or adversary.
In the ideal type of inquisitory tribunal, the judge supervises and even
himself carries out much of the process of calling evidence, and examin-
ing and cross-examining witnesses; he takes the initiative in establishing
what happened. A basic theory of this type is characteristic of most
systems which have been influenced by Roman and Canon law. In the
adversary system, the main responsibility for calling, adducing, and

30 In Hobson v. Hobson, (1942), 59 W.N. (N.S.W.) 85, Bonney, J. gives an
excellent account of blood groups as evidence of paternity. This was cited
with approval in R. v. Jenkins, ex parte Morrison, [19493 V.L.R. 277 (Sup. Ct.)
(though there was much expert evidence in the latter case also). Since 1942,
much more has been discovered about the matters discussed by Bonney, J.,
and it would be dangerous for a court today to rely on his account of them.

31 Popper, op. cit., n. 3, esp. at pp. 49-52.

McGILL LAW JOURNAL

[Vol. 18

testing evidence is with the parties; the judge has no independent power
of calling evidence, interferes little with the conduct of the parties or
their representatives when handling evidence except by way of excluding
inadmissible material, and regards his task as judging on the basis
of what he is told, not on a basis of the ‘actual’ facts. This approach
is characteristic of most systems derived from English common law.3 2
This is an accurate general description, the force of which is
not lessened by the fact that no trial system is completely “pure”;
the common law trial has some inquisitorial features, and those
European trial systems which have been influenced by Roman
and Canon law some adversarial ones. Nevertheless the basic ad-
versarial approach is distinctive of the common law trial, and is
supported by the vast majority of practising lawyers in common
law jurisdictions. Even the late Judge Frank, who bitterly con-
denned some features of our trial system, regarded them as excesses
or excrescences on a basically sound approach. In support of his
view that the adversary approach is sound, he cites Macaulay’s
remark that we obtain the fairest decision when two men argue,
as unfairly as possible, on opposite sides, for then it is certain
that no important consideration will altogether escape notice.8 3
The source of Macaulay’s remark is not given, and he may well
have been discussing the presentation of argument. In that event,
what he was saying would have been true enough. But totally
different considerations apply when we are dealing with the ad-
duction of evidence.

Imagine a case of a scientist asked to decide which of two rival
theories, each propounded by an eminent scientist, is to be con-
sidered correct. If he is presented by the two rivals with all the
relevant data, and then given the arguments of each, presented as
strongly as possible, he might be able to perform the task. Even
if he thought that neither rival was completely in the right, he
could construct his own theory by borrowing elements from each
of those already put before him. We might liken this case to that
of an appellate court deciding between rival advocates.

But no scientist would dream of making a decision such as
that which I have just posited, if he knew (a) that neither of
the rival theorists was bound to put forward all the data in his
possession –
indeed, that each would regard it as proper to suppress
any “inconvenient” or inconsistent observations of whose existence
he knew and (b) that he, the adjudicator, would be precluded

32 G. Sawer, Law in Society, (Oxford, 1965), at pp. 72-73.
33 J. Frank, Courts on Trial, (Princeton, 1949), chapter 6.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

from suggesting or requiring the elicitation of any additional data
that might prove critical. Indeed, he would ordinarily expect to
perfect his theory or make his decision by pursuing every line of
enquiry which he thought might throw light on the matter.

It is precisely because the two factors just mentioned operate
in the ordinary adversary trial that the latter must be regarded
as basically unscientific in approach, and unsound. The judge –
the court, that is to say –
is supposed to be deciding the case
by applying the law to the “true facts”. In saying this, I realise
as fully as does anyone else that the “true facts” can never be
ascertained (since the past cannot be completely recalled), and
that the reality of a common law trial is that it is a battle between
the parties rather than a search for truth. But we do not advertise
our courts as substitutes for boxing rings or jousting fields. We
hold them out as being temples of justice engaged in the search
for truth, and I do not believe that society would tolerate their
continued existence on any other basis. If this is accepted, then
we may recognise that they do not at present in fact search for
the truth, but we ought at least to try to improve their actual
performance so that it more nearly resembles the ideal.

At a common law trial there is no real reason for supposing that
the truth will emerge; for the adversary system is designed so as
to permit the parties to conceal it from the court. If one or other
party is aware of inconvenient evidence which casts doubt on –
nay, refutes –
the story which he puts forward as a basis for
adjudication, he is perfectly entitled to withhold it if he can. He
takes the risk that the other party may triumphantly produce it;
and if that happens, the court will learn of it. The other party may,
however, be unaware of the existence of the evidence, in which
case the court will never learn of it. This cannot be answered by
asserting confidently, with Macaulay, that it is certain that no
important consideration will altogether escape notice. Such an
assertion confuses an incentive to obtain contradictory evidence
with the capacity or ability to obtain it.

Nor can my criticism be met by saying that the point may be
noticed by the court even if the opposing party overlooks it. In
such an event the second unscientific hobble placed on the court
in our system comes into action; the court cannot call for evidence.3 4

34 Titheradge v. R., (1917), 24 C.L.R. 107 (High Ct. of Australia), esp. per
Isaacs and Rich, JJ., at p. 118. Moreover, severe restrictions are placed on
the presentation of further evidence (in rebuttal) by a party who has closed
his case: Shaw v. R., (1952), 85 C.L.R. 365. These decisions, although given
in criminal cases, also apply, as the judgments show, to civil cases.

McGILL LAW JOURNAL

[Vol. 18

No doubt the court can suggest very strongly to counsel that certain
evidence ought to be presented. But if, for tactical or other reasons,
counsel declines to present it, the court must go without it.

I need not give examples by way of elaboration, for what I
have been saying is within the knowledge of every practising law-
yer. I wish, however, to add one point, which is that the deficiencies
of our adversary system operate most harshly and dangerously in
criminal as opposed to civil trials and wherever reliance is placed
on circumstantial evidence.

I need say little about the special problem of criminal trials.
It is surely obvious that even the most favourable things that can
be said in support of the adversary system require a presupposition
that the contending parties are more or less on a par as regards
access to sources of evidence. That presupposition is plainly untrue
where, as in a criminal case, the individual is pitted against the
state. It is not simply a matter of financial resources, though of
course many criminal defendants are poor. It is rather that the
police have in practice (whatever the theory) quasi-coercive powers
of obtaining statements from witnesses, as well as many more
investigators.

Circumstantial evidence poses a different problem. Let me say
at once that I agree with the view that it is, if of high quality, far
more persuasive and far less open to error than testimonial evi-
dence 5 But just as the latter possesses defects of which our courts
take no account (as I shall later argue) so does circumstantial
evidence often pose a problem that is overlooked.

The proper use of it can be regarded as a paradigm case of
scientific method. The investigator, being confronted by a problem,
forms a hypothesis to account for it; in this stage the process is
outside logic; Medawar (whose account of scientific method 8
I
am following) terms it intuitive. But having formed the hypothe-
sis, the investigator subjects it to rigorous testing and criticism.
He makes use of logic by asking what consequences flow from
his hypothesis and then takes steps to find out whether what
follows logically is indeed the case. At this stage it is not enough
to say that the hypothesis accords with the data already in exist-
ence, nor is it enough to look for merely confirmatory data (re-
member Popper’s remark that we can always find confirmatory
evidence if we look for it). The investigator must endeavour to

35 1 Wigmore on Evidence, (Boston, 3rd. ed., 1940), secs. 25, 26.
36Medawar, (1969), op. cit., n. 4, esp. sec. III. See also: Medawar, (1967),

Hypothesis and Imagination, op. cit., n. 3.

No. 21

IMPLICATIONS OF SCIENCE FOR THE LAW

find consequences that would render his hypothesis false, and it
is when he fails to do so that he can place confidence in his hy-
pothesis.

A really good police investigation would follow this procedure.
In practice, however, what happens is – more often than not, I
believe –
that the investigator pins his faith to the confirmatory
data that he has and looks for more of the same, and disregards
or explains away any apparently contradictory facts. Though this
is an understandable human failing, the result is that many trials
begin with a theory, based on circumstantial evidence, that has
not really been tested with a view to falsification.

Let us, however, assume that the trial starts with an apparently
sound prosecution hypothesis. At this point difficulties come thick
and fast. The investigatory process is at once split up. The theory
is presented by the prosecuting counsel, not by the investigator
who conceived it. The witnessess give evidence which is intended
by the prosecution to confirm the theory. There is, however, no
guarantee that the conclusions drawn by the jury from this evi-
dence will be the same as those drawn by the original investigator.
Moreover, the defence may now present further evidence which
tends to falsify the original hypothesis. At the end of all this the
jury draw their conclusion, but all we know from them is the
result. We are precluded from seeing what hypothesis they acted
on and hence from asking whether it was in any way falsified by
the data.

It is possible that in this way the jury may arrive at a true
and just conclusion. My point is that we have no reason to assume
that they will. To be sure, they will have been instructed that they
must acquit if there is in their view a reasonable doubt as to the
accused’s guilt.37 But to carry out such an instruction requires not
merely a willing and conscientious mind (which we ought to at-
tribute to each juror) but a trained mind. It may easily happen
that the jurors will think that pieces of evidence A and B point
inexorably to guilt, and that so do pieces C and D, and neverthe-
less fail to realise that hypothesis X, for which A and B account,
is quite inconsistent with hypothesis Y, accounted for by C and D.
The point was tellingly illustrated in the case of Colin Campbell
Ross, who was hanged in 1922 for the felony-murder of a young
girl. On his appeal to the High Court of Australia against his con-
viction, Isaacs, J. said, in the course of a long dissent, that:

379 Wigmore, op. cit., n. 35, sec. 2497.

McGILL LAW JOURNAL

[Vol. 18

The Crown placed before the jury a mass of evidence, but in that mass
there were three principal pieces of testimony: the Matthews confession,
the Maddox confession and the Harding confession. On certain crucial
points they were absolutely inconsistent. But the Crown left to the jury
the choice, inter alia, of accepting any of them as having been made
and as being true; the whole matter now centres around this one vital
question: What is the true position, supposing the jury believed that Ross
did make the Harding confession, and that every word of it was true?
Now, to my mind, if the Crown arguments are constantly tested by this
one question, “Suppose the Harding confession was accepted as true”,
all difficulties disappear, because, being itself a denial of intention to
kill, it at once excludes other evidence relied on to show a contrary state
of facts. It is only when that question is lost sight of, and it is conjectured,
that the jury did an impossible thing, namely, accepted the Harding
confession as the true state of facts, namely, that murder was not intended,
and at the same time drew from other evidence an inference entirely
inconsistent with it, namely, that murder was intended, that the verdict
can be sustained.3 8
This powerful argument was rejected by the rest of the Court.
In their views, the trial judge had correctly instructed the jury
as to what constituted murder, and the jury were entitled to apply
his instruction to so much of the evidence as they chose to accept;
they “were at liberty to believe as much of or as little of the
[Harding] confession as they choose”. 9 This may be technically
correct,40 but it is surely most unsafe to assume that having arrived
at their own construction of what the prisoner had in fact done,
they were capable of correctly applying to that version of the facts
(which had not until then been advanced) a general direction as
to the law of murder.

A rather different species of the same generic problem –

en-
suring that the hypothesis actually acted upon has been adequately
is exemplified by such cases as Russel 41 and Bullock.42
tested –
In these the trial proceeded with the prosecution offering a specific
hypothesis as to how the accused had committed the crime charged,
which hypothesis the accused flatly denied. In their deliberations
the jury apparently concocted a totally different hypothesis and
sought further instructions from the judge, in the light of which
they in each case convicted the accused. The conviction was upheld

38 Ross V. R., (1922), 30 C.L.R. 246, at pp. 258-259.
39 Ibid., at p. 253.
40 The rule is that the criterion of “no reasonable doubt” applies to the
whole of the issue, not to specific aspects of the evidence. See: 9 Wigmore,
op. cit., n. 37.

42 [1933] Vict. L.R. 59.
42 (1954), 38 Cr. App. R. 151.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

on appeal after the further instruction had been adjudged to be
good law. Again, this may be technically correct; but it will be
observed that in neither of these cases had the jury’s hypothesis
been subjected to any testing whatsoever against the available data,
and still less against any additional data that might have been
obtained by recalling some one or more of the witnesses.

The foregoing discussion may be summarised by saying that
our adversary system of trial is utterly unscientific in method,
and unlikely to help in the ascertainment of the facts which are
the foundation of every legal dispute. The inquisitorial system
follows the general lines of a scientific investigation and is much
to be preferred. There is little to prevent us from changing our
present system. Of course, the latter places a premium on tactical
advocacy, and change is accordingly likely to be resisted by those
who make an excellent living by their skill in using tactics. Also
the word “inquisitorial” conjures up, somewhat unjustifiably, dark
pictures of the Star Chamber and the Spanish Inquisition; it would
be better to use another word, such as “investigatory” or “initia-
tory”. Moreover, the adoption of such a system would involve
the disappearance of jiuy trial in its present form (since the mental
operations of the investigator-decider must be open to scrutiny
and criticism). It would not, however, necessarily involve the com-
a step which I would not wish
plete disappearance of juries –
to see taken. Juries could remain as assistants or assessors to the
judge in the task of investigating and settling the facts, and they
could be given a decisive voice in the eventual disposition of the
case on the basis of those facts.

I now turn to the basic postulates of the law of evidence. The
common law trial system is distinctive in the emphasis it gives to
the testimony of witnesses actually present in the court. This em-
phasis reflects the ubiquitous operation of the hearsay rule, and
that rule, in its turn, is derived from an insistence that the most
reliable evidence obtainable is that of witnesses who will recount
what they have seen or heard at some past time, under the twin
guarantees of an oath and a liability to be cross-examined.4 3 In
short, underlying much of our system of evidence are two assump-
(1) that a person is more likely to tell the truth if he
tions –

435 Wigmore, op. cit., n. 35, sec. 1362. Wigmore regarded the requirement
of an oath as of secondary and minor importance, but his reasoning in the
section referred to lacks factual support at the critical point. He realises
this objection, but brushes it aside with the remark that the factual support
is not obtainable.

McGILL LAW JOURNAL

[Vol. 18

that whether a person
has sworn an oath to do so, and (2)
is or is not telling the truth can best be tested by cross-examination.
These assumptions could easily enough be empirically tested.
So far as I know, however, no real effort has been made to test
either of them. Perhaps one reason for this failure to even attempt
to verify our hypotheses springs from a fear that they would be
shown to be incorrect, and that we should have to reconsider the
very foundations of our system.

It may have been true at some time in the past that the taking
of an oath acted as a powerful spur to a witness to tell the truth.
But as I noted at the outset, the influence of religious beliefs on
the conduct of our daily lives has greatly declined in recent dec-
ades, and it is at present reasonable to suppose that most witnesses
who take an oath are quite unworried by the possibility of super-
natural sanctions. At the same time I do not doubt that the oath
has the effect of making some witnesses (for example, those with
strong religious beliefs or consciences) hesitant and worried and
perhaps less helpful than they otherwise might have been. My
point, however, is that on these matters we are content to guess
and assume when we could quite easily take some positive steps
to find out.

So also with cross-examination. Wigmore describes it as “the
greatest legal engine ever invented for the discovery of truth” and
says that “the belief that no safeguard for testing the value of
human statements is comparable to that furnished by cross-exami-
nation.., has found increasing strength in lengthening experience”. 4
He quotes numerous examples in support of his thesis, though the
value of these is lessened when we once more recall Popper’s re-
mark 45 that it is always easy to find confirmations of a hypo-
thesis if you look for them. And revealingly, he adds that “it may
be that in more than one sense it [cross-examination] takes the
place in our system which torture occupied in the mediaeval system
of the civilians”. 40

The last remark suggests that there may have been some con-
fusion in Wigmore’s thinking on this matter. And I believe that,
in truth, he is confusing two distinct things. There is, first, the
scientific method of testing an hypothesis by critical examination.
In this respect we may regard cross-examination as scientific in
approach, for it allows the evidence of a partisan witness to be

44 5 Wigmore, ibid., sec. 1367.
45 See text, p. 173 reference to point 1.
46 5 Wigmore, op. cit., n. 35, sec. 1367.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

critically tested. But we may notice that here “cross-examination”
is simply a synonym for “critical interrogation”, and that it could,
as such, be carried out quite as effectively by an active, investi-
gating judge as by a hostile, opposing counsel.

It is this hostile questioning by an opposing counsel bent on
destroying the witness’ testimony in chief that is the other and
more common entity described by the term “cross-examination”.
And this is far removed from a mere critical questioning of the
witness. For the latter is impartial in approach; the tester is seek-
ing the truth, even if it be what the witness has claimed it to
be. The cross-examiner, however, is not seeking the truth; he is
seeking to win the case for his client by destroying the witness’
credibility even if the witness has told the truth. And he is aided
in this task by being allowed to badger and bully the witness, to
seek to trap him, to try to wear him down by repeated questioning,
to sneer at and ridicule him, to deprive him of any opportunity to
explain apparent discrepancies in his answers –
I need not length-
en the catalogue. Anyone who has seen a “powerful cross-exami-
nation” will readily appreciate why Wigmore compared it to torture.
We might also recall the howls of protest that are raised when
another type of interrogation –
for example, the questioning of
a suspect by a police officer –
is carried out with the same tech-
niques.

While I do not doubt the value of testing a witness’ recollectibns
by careful, calm, critical questioning, I am inclined to believe that
cross-examination is a quite different process which is likely to
produce a quite different effect. I believe that in the majority of
cases it is likely to produce confusion and error rather than truth.
In a small minority of cases it may help to reveal the truth, and
in yet other cases it will have no real effect other than to leave
the witness with an abiding dislike of courts and court procedures.
These are no more than my own beliefs. They, like the opposing
“official” beliefs of the law, could be tested empirically. It would
not be easy to devise a really careful test, but it could be done.
Once again, so far as I know, there has been no real effort to
carry out any such test.

In the meantime, the belief in the need for an oath and our
opportunity to cross-examine continues to dominate our system
of evidence and forces us to give pride of place to the witness in
court. We rely very largely on what he today tells the court about
what happened at some time in the past. So we should be led to
ask: is this reliance justified?

McGILL LAW JOURNAL

[Vol. 18

As Wigmore points out, the notion of taking a witness’ court
statement as evidence of a past occurrence impliedly attributes
to the witness the three processes of observation, recollection, and
communication. The witness must have observed the event in ques-
tion and in so doing have received some impression of it in his
mind; he must now have a recollection of that impression; and
he must communicate his recollection of that impression to the
court.47 We need not here concern ourselves with the problem of
communication, for it is plain that if communication is to be of
use, it must have been preceded by accurate observation and re-
collection.

“The notion of Observation”, says Wigmore, “is that the external
event has in some way or other impressed itself on the witness’
senses … This impression … should adequately represent or cor-
respond to the fact itself as it really existed or exists … Again,
the function of Recollection is to recall or reproduce the original
impressions of observation…”48 This analysis seems accurate; but
it should at once lead us to ask whether such observation or re-
collection can or does ever occur.

On this point there is a wealth of scientific information avail-
able to us. Technically speaking, it must be regarded as provisional,
in the sense that (like any other scientific material) fresh data at
some future time might require an abandonment or modification
of what is believed at present. But the material I am referring
to has been known for many years, and has withstood repeated
and critical testing. If anything, the future seems more likely to
bring an extension of what is at present thought rather than a
repudiation of it. And consequently the law can (for the time
being) safely act upon the present understanding.

I do not wish to give the impression that I am the first to raise
these matters. On the contrary, references to some of this scientific
material are to be found in the law reviews and in some legal
texts.4 9 But for the most part it is used to warn lawyers to be
on their guard against possibly faulty observation or recollection
in specific cases.

This understates the position. In truth, we can rather say that
no witness can observe or recollect in the manner suggested by
Wigmore. The reason for this lies in the very structure of the

47 2 Wigmore, op. cit., n. 35, sec. 478.
48 Id.
49 E.g. J. Marshall, Law and Psychology in Conflict, (Indianapolis, 1966).

No. 2I

IMPLICATIONS OF SCIENCE FOR THE LAW

human brain, sensory organs, and central nervous system. Our
eyes are not cameras which convey accurate pictures of all that
exists in front of them to the brain, which then records the result
like a videotape to be stored and played back when the appro-
priate button is pressed. The true situation is almost the reverse
of that process. The eye consists of numbers of light-sensitive
cells which respond in different ways to external stimuli. The optic
nerve conveys the reactions of these cells to the brain which pro-
cesses, selects, combines, and interprets them so as to produce
a “mental picture”. In so doing it rejects what is “unimportant”
or “uninteresting”. In brief, the act of vision is a creative, not
a recording, act.50

Hence it is correct to say that we do not see and cannot see
what is “there” to be seen. We see, rather, so much of what is
“there” as may be of importance to us (for example, to enable
us to survive), or of interest to us, or in accordance with our
expectations of what will be “there”. If what is “there” is un-
important, or uninteresting, or not something we are expecting
to see, we shall not see it. Again, the detection of an object against
a similar background, or of something that would ordinarily be
moving when it has “frozen” to avoid observation, is difficult be-
cause the brain has, so to speak, to work against the structure
of the eye; the eye is built so as to perceive dark against light,
to perceive movement against a still background, and so on. We
may summarise this by saying that, if by “faulty” observation
we mean, with Wigmore, observation that does not correspond
to the material that exists to be observed, all observation is faulty.
So also with other sense impressions –
touch, taste, smell and
hearing.

If we pause to think about the matter, the selective and creative
nature of our handling of sense impressions from the external world
is something we are all familiar with. Who has not engaged in a
conversation at a party, conscious of a mere buzz of noise from
the other guests in the room, only to find himself suddenly dis-
tracted by hearing his name mentioned by a guest in the opposite
corner? This phenomenon is often discussed in the literature as
the “cocktail party” effect, and it illustrates the way in which we
reject and disregard many or most of the sound waves impinging
upon our ear drums but select out those of importance or interest
to us.

50 p. Nathan, The Nervous System, (1969).

McGILL LAW JOURNAL

[Vol. 18

The selectiveness of perception is, however, only part of the
infirmity inherent in placing reliance on the recollected observa-
tions of a witness. For when we investigate the phenomenon of
memory, we find a similar selectiveness. The act of remembering
involves the storage and retrieval of information, but it is far
more complicated than that. In a series of elegant experiments
carried out in the earlier part of this century, Bartlett showed
that:

… remembering is not the re-excitation of innumerable fixed, lifeless and
fragmentary traces. It is an imaginative reconstruction, or construction,
built out of the relation of our attitude towards a whole active mass
of organised past reactions or experience, and to a little outstanding
detail which commonly appears in image or in language form. It is thus
hardly ever really exact, even in the most rudimentary cases of rote
recapitulation, and it is not at all important that it should be so. 1
Bartlett produced a theory to account for the results of his
experimental observations. The theory has been subjected to some
criticism, and anyway it does not concern us here. The observations
that led to it are what is important, and more recent data have
confirmed them. Since Bartlett wrote, much work has been done
by neuro-physiologists, but it is fair to say that the mechanism of
memory is still not fully known. Nevertheless, we find Nathan, a
research neurologist, summarising the results of present knowledge
in much the same was as Bartlett:

A remembered scene is not the same as the original scene. It is pale
and weak in all respects. The accompanying emotion is far less, much
of what was first experienced has dropped out of the remembered scene,
much else has become changed or distorted and a great deal has been
forgotten… Remembering is similar to imagining a scene;
in some
respects, remembering is re-imagining. The memory is a reconstruction,
based on the original experience.5 2

Later he reminds us that “a conscious effort to look for the past
experience is not always the best way of finding it.” 13

The effect of all these teachings is surely clear enough. The
picture of the witness as a passive perceiving, recording, and
recalling machine – which is how, according to Wigmore, the
common law system of trial views him –
is totally false. It is
not hard to see why this should be so. The common law theory
was built up long ago, well before any scientific investigation of
perception and memory was attempted. On that theory an enormous

51 F. Bartlett, Remembering (reprinted ed., 1950), at p. 213.
52 Nathan, op. cit., n. 50, at p. 338.
53 Ibid., at p. 348.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

the whole detailed system of rules of evidence –

superstructure –
has been erected. The foundations have disappeared from view,
and so we fail to realise that with the passage of time they have
crumbled away.

Biologically, perception and memory enable us as living beings
to cope with the present by appropriate behaviour. Their purpose
is not to enable us to relive the past. But a trial is concerned
with bringing the past to life. To rely on the present testimony of
witnesses to achieve that end is to court disaster.

There exist, however, many means of reliving the past in the
present. A tape recorder will reproduce in the present a sound
from the past; similarly, a photograph or a videotape will re-
produce a sight from the past. A contemporary document will
time, and
reproduce
although the imperfections of faulty observation and recollection
will affect its reliability, they will probably be fewer and smaller
than those which occur when the maker of the record tries today
to recall the event.

its maker recorded at the

today what

I would think, therefore, that a scientific approach to the re-
construction of a past transaction would lead us to prefer the
mechanical record to the human, and the contemporary statement
to the present recollection of a witness. Such an approach is
almost the exact opposite of what occurs in the courts.

In saying this I have not overlooked the obvious objection
that tape recordings can be tampered with, contemporary documents
altered, and so on. Of course they can; and the possibility that
such falsification has occurred should not be overlooked in a given
case. But the danger can be guarded against, and it is of minor
importance when we realise that reliance on the present testimony
of witnesses contains an inbuilt falsification process that cannot
be removed.

The foregoing discussion leads naturally to my third point –

the incorrect portrait of man which underlies many of our sub-
stantive rules of law. Here I am concerned with a different aspect
of human functioning. I referred earlier to Popper’s view that Freud’s
basic theory of psycho-analysis could not, because of its lack of
susceptibility to being tested, be accorded the status of being
scientific. I indicated my agreement with that view, and it is now
appropriate for me to indicate my agreement with Popper’s further
remark that:

… this does not mean that Freud and Adler were not seeing certain things
correctly: I personally do not doubt that much of what they say is of

McGILL LAW JOURNAL

[Vol. 18

considerable importance, and may well play its part one day in a psycho-
logical science which is testable.f4
Whether or not he was the originator of the idea, Freud must
surely be credited with having made us aware that the human
mind functions at different levels. As Nathan puts it, “when we
are conscious, so much of our mental and psychological activity
is in fact unconscious. We make use of conscious effort only in
small parts of our mental activities.” 55

The reader may well say that there is nothing startlingly new
about this observation. What is startling, however, is the realisation
that the vast majority of our daily actions result from the un-
conscious activity of the brain, and that this is a biological necessity.
Successful living depends in part on an ability to relegate what
begins as conscious action to the domain of the unconscious. The
piano student begins by having to direct his conscious mind to
getting his fingers to depress the keys correctly, and as he progresses,
he relegates much of the activity of “playing the piano” to an
unconscious level and frees his conscious mind to concentrate on
the more important process of interpreting the music. This is the
process of acquiring a technique or skill, and it involves, as a
rule, conscious repetitive actions, until the unconscious can “take
over”. So also with the baby learning to pick up objects and to
walk, and with the adolescent learning to drive. The brain processes
involved are still in the very early stages of investigation, though
it has been repeatedly shown that the activity of the brain is
ceaseless. Nor should we overlook the mysterious ability of the
skilled performer to bring what would ordinarily be relegated to
the unconscious level temporarily into the realm of complete con-
sciousness, so that he can modify what he is doing.

The relevance of this can be simply stated. The rules of law
pay no attention to the importance of unconscious activity. They
assume that we all function continually at a single level of complete
consciousness and fault us when we have failed to do so.

It would take a series of volumes on the various branches of
the law to illustrate this thesis, and I do not propose to lengthen
this essay by attempting the task. 0 It is enough to remind the
reader of the current vogue in criminal law for insisting on purpose,
foresight, and the assumed behaviour of the “reasonable man” as

54 Popper, op. cit., n. 3, at p. 37.
55 Nathan, op. cit., n. 50, at p. 338.
56 Or to add footnote references to each point discussed, since the doctrines

mentioned are known to every lawyer.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

the measure of liability, with all the distortions that result there-
from; how, for example, can we sensibly demand of a man who
is attacked and defends himself that he should use no more force
than is reasonably necessary? In doing so are we not in reality
asking that he should direct his attention to measuring the precise
degree of force being used by the aggressor? And if he attempted
to do so, would he not greatly reduce the likely success of his
defence?

Again, much of the law of tort ranges around negligence and
the concept of foreseeability. To be sure, in certain stereotyped
classes of action, such as motor accident litigation, we turn to
specific questions such as whether the defendant driver was keeping
a proper lookout. But even then, the notion seems to be that he
should be constantly considering at a conscious level every possi-
bility of danger, although if every driver were to do this there
would first be a large crop of accidents, following which all traffic
would grind to a halt. In less stereotyped cases the notion of
foreseeability is carried to absurd lengths, and courts demand
of litigants that they should have directed their minds to (fore-
seeable) events which would never occur to the ordinary person’s
mind.

Then again, there are the many cases in the fields of contract
and commercial law in which the courts, faced with an agreement
that makes no provision for an event that the parties never con-
sidered, decline to go beyond the agreed terms, remarking that
since the parties could have made some provision on the matter
but failed to do so, the court will not make one for them. In
effect, the courts in such cases penalise the parties for making
unconscious assumptions about the expected continuity of con-
ditions. Allied to these are the cases in which the courts assume,
in making their decisions, that in their everyday dealings people
read the conditions endorsed on tickets, dockets, and the like, that
they are capable of correctly understanding and interpreting what
they supposedly read, and that it would be possible for them to
stipulate for and obtain changes in the conditions if they did not
like the effect of them.

These are a few examples. The reader will doubtless find others
occurring to his mind, once he directs his attention to the point.
The assumption that each of us functions as a constantly thinking
and foreseeing being, addressing his mind consciously to every
aspect of every transaction that he encounters every day, pervades
the whole of our law. It is a false assumption. Necessarily, most
of our daily activities are carried out automatically by unconscious

McGILL LAW JOURNAL

[Vol. 18

mental activity, and even where we think about a given matter
we continue to make, without conscious thought, numerous as-
sumptions as to surrounding circumstances. It may be possible, with
arduous effort, to bring some of this unconscious activity under
conscious control, but few of us have been trained for the task
or are capable of being successfully trained for it. It is probably
desirable that this should be so, as any large-scale attempt to
eradicate unconscious mental activity would rapidly bring normal
life to a stop.

Surely, then, our rules of law should be bottomed on the actual-
ities of human mental life instead of on some impossible ideal.
The law should deal with and regulate the lives of human beings
as they really exist. At present it endeavours to fit humanity into
a Procrustean bed of its own devising.

Lastly, I turn briefly to the rule of stare decisis. In recent
years it has been somewhat relaxed, and in its stricter forms it
operates in very few courts (the Court of Appeal in England is one
of them 57). But although the rule of stare decisis no longer retains
its full force, it still exercises a potent influence on judicial practice.
Courts which assert that they have power to overrule their previous
decisions add to that assertion conditions to hobble themselves,
such as that the power must be exercised with great caution or
only in very special circumstances. The emphasis on the sanctity
of an earlier decision remains.

The attitude of the courts in our common law system is thus
one which is basically dogmatic and authoritarian. As such it is
the opposite of a scientific attitude, which, although not repudiating
received beliefs merely because they are received, is determined
to approach them in a critical spirit, to test their validity, and to
reject them if, as a result of the test, they are found to be unsound 8
There are, of course, eminent lawyers who would dispute this
point. Indeed, very recently, two members of the Court of Appeal
in England have asserted (following in this regard earlier legal
authorities) that law is a science 9 The basis of this assertion is
that just as such sciences as physics and chemistry are concerned
with the measurement and analysis of physical substances, so the
law is concerned with the analysis and measurement of reported
decisions.

57Barrington v. Lee, [1971] 3 W.L.R. 962 (C.A.).
5 8 Popper, op. cit., n. 3, at pp. 49-52.
59Incorporated Council of Law Reporting v. Attorney-General, [1971] 3

W.L.R. 853 (CA.), at pp. 865 (Sachs, L.J.), 874 (Buckley, L.J.).

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

The analogy is on its face a poor one; there is a world of
difference between analysing and measuring physical substances
and analysing and measuring words. But of much greater importance
than the poverty of the analogy is the total misunderstanding of
the nature of scientific activity which the judges display. Scientists
are not engaged in collecting “facts” for analysis, in the hope that
some “law” will turn up in the process. They are engaged in
formulating hypotheses to explain problems that interest them, and
in the rigorous criticism and testing of those hypotheses. As Medawar
puts it, “scientists are building explanatory structures, telling stories
which are scrupulously tested to see if they are stories about real
life.” 60

A case recently decided by the High Court of Australia”‘ illus-
trates just how different is the procedure of a court. A succeeded
against B at the trial of an action, and B appealed to the Court
of Appeal in New South Wales. He was ordered on 10th February
by that Court to file and serve appeal books on or before 31st March.
Owing to delays by the law stationers, only four copies of the ap-
peal book were ready by 31st March, and these were filed on that
day. Copies to serve on A’s solicitors were not ready until 6th April,
when they were served. The order of 10th February had been
made in a form which stated that if the filing and service of
the appeal books did not take place on or before 31st March
the appeal was “to stand dismissed for want of prosecution”. Never-
theless, in September the Court of Appeal ordered that the filing
and service which had taken place was a sufficient compliance
with its order of 10th February, thus allowing the appeal to proceed.
From this latter order A appealed to the High Court. That Court
held (Gibbs, S. dissenting) that the appeal stood effectively dis-
missed on 31st March and that neither the Court of Appeal, nor
for that matter any other court, had an inherent jurisdiction to
revive a dismissed appeal.

The Court reached this conclusion partly on the basis of au-
thority. Most of the latter was technically of persuasive status only,
but there was also a binding precedent in the form of an earlier
unreported decision of the High Court. As Gibbs, J. pointed out,
the persuasive authority was susceptible of differing interpretations,

GOMedawar (1967), Hypothesis and Imagination, op. cit., n. 4. See also:
(1967), Two Conceptions of Science, op. cit., n. 4 and Medawar
Medawar
(1969), Induction and Intuition in Scientific Thought, for repudiations of the
view that the essence of scientific activity lies in measurement, analysis, etc.

61 Bailey v. Marinoff, (1971), 45 A.L.J.R. 598.

McGILL LAW JOURNAL

[Vol. 18

and the earlier High Court decision was not only technically dis-
tinguishable, but had also been given without adequate argument
by one of the parties (who appeared in person) and without calling
on counsel for the other party for any argument at all on the point.
Nevertheless the judges of the majority saw no sufficient reason
(note the mode of approach) for reconsidering the earlier decision
of the Court.

These judges were of course not content to rest there. They
made it plain that they thought the reason for both the earlier
and the instant decision a sound one –
namely, that a court
which allowed the reinstatement of a proceeding which it had finally
disposed of would not be promoting the due administration of the
law or of justice. This proposition, if I may use Medawar’s telling
phrase,62 was made not by reasoning but by asseveration. Yet it is
a proposition which in some respects is capable of being tested.
The courts of the United States, I believe, allow the exercise of
powers such as that denied by the High Court, and do so without
undue harm to the administration of justice. Moreover, if it be
said that there are some undesirable features which accompany
the exercise of such a power in the United States, one could go
on to enquire whether those features are inevitable accompaniments
of the power’s exercise or are capable of being avoided. As for
the promotion of justice, one perhaps cannot test the matter in the
same way (for justice appears in different garbs to different eyes);
but I leave it to the reader to decide whether decisions such as
this promote either justice or respect for the courts.

The case illustrates the two main respects in which our judicial
method, in respect of the use ,of precedent, is the antithesis of
scientific method. Firstly, there is the dogmatic reliance on past
authority as virtually unquestionable. This contrasts with the scien-
tific attitude of continuing critical evaluation and modification of
received notions. The courts’ attitude can in part be explained by
a decent respect on their part for their legislatures as proper
instigators of major changes. But it is well known that in many
respects the legislatures do not expect to be called on to intervene
they leave the courts to get on with their own business of
–
promoting the orderly development of the law and its adaptation
to contemporary conditions. The courts’ own attitudes, however,
prevent them from satisfactorily discharging the latter task.

Secondly, the rule enunciated in the case is grounded upon its
supposed effects, or upon the effects which a different rule would

62Medawar (1969), op. cit., n. 4, at p. 6.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

supposedly have. But when courts have once reasoned in this way,
they continue to reiterate the reason without attempting to in-
vestigate whether the desired results in fact occur. For example,
more than 20 years have elapsed since a number of United States
courts 13 adopted a rule (given constitutional force 10 years ago 64)
prohibiting the use in criminal prosecutions of evidence obtained
by means of an unlawful search or seizure. The avowed purpose
of the rule was to deter the police from resorting to unlawful
practices. The number of cases still reaching the courts in which
the rule is applied suggests that the police are not in fact being
deterred. But there is virtually no sign of any attempt by the courts
to reconsider the rule in the light of its actual effectiveness. 5

In short, the courts totally ignore the possibility of making
neatly defined by Medawar as “the control of
use of feedback –
perfomance by the consequences of the act performed” and de-
scribed, also by him, as a “general stratagem that underlies almost
all regulative process or processes of continuous control”. 6 Feed-
back is an essential feature of scientific method. The scientist devises
a hypothesis, draws inferences (logically) as to its consequences
and carries out tests to ascertain whether the consequences of the
hypothesis in fact occur. If they do not, he modifies the hypothesis
to take account of his new data and starts the cycle anew. In
contrast, the courts frame a hypothesis by reference to its supposed
consequences, and convert the hypothesis into a “law” with dis-
dainful disregard of the question whether the consequences ever
could or do occur. Yet there is ample room for the incorporation
of feedback into a legal system. It is the spirit of scientific enquiry
that is lacking.

Indeed, the whole system of stare decisis rests on assertion
rather than critical investigation. Courts assert that the system is
essential to produce certainty, like treatment of like cases, and
other desirable features of a system of legal control. Yet books
have been written which amply demonstrate that the system pro-

03 1n 1949 only a minority of jurisdictions had adopted the rule discussed
in the text. See: Wolf v. Colorado, 338 U.S. 25; 93 L. Ed. 1782 for a summary
of the position. In 1955 the Supreme Court of California reversed its earlier
decisions and adopted the exclusionary rule: People v. Cahan, 44 Cal. 2d. 434;
282 P. 2d. 905. In so doing it gave a lead which was rapidly followed in other
jurisdictions.

64Mapp v. Ohio, (1961), 367 U.S. 643; 6 L. Ed. 2d. 1181.
65 Burger, CJ. criticised the rule on the ground of its, alleged ineffectiveness
in Bivens v. Six Unknown Federal Narcotics Agents, (1971), 29 L. Ed. 2d. 619.

06 Medawar (1969), op. cit., n. 4, at p. 54.

McGILL LAW JOURNAL

[Vol. 18

duces none of these effects. They are ignored. So, too, is the ex-
perience of the European systems of law that operate without a
rule of stare decisis.

In summary, then, the rule of stare decisis –

the very heart
of our legal system –
is irreconcileable with the scientific approach.
So long as we retain it, albeit in a weakened form, the claim that
law is a science must be treated with derision.

IV

In his elegant monograph Elizabeth and Essex,67 written during
the earlier part of this century, Lytton Strachey contrasted the
procedure which led to the execution of Dr. Ruy Lopez for treason
with the improved methods current in his own time. He said that:
The true principles of criminal jurisprudence have only come to be
recognized, with gradually increasing completeness, during the last two
centuries; the comprehension of them has grown with the growth of
science – with the understanding of the nature of evidence, and the
slow triumph in men’s mental habits, of ordered experience and reason.
No human creature can ever hope to be truly just; but there are degrees
in mortal fallibility, and for countless ages the justice of mankind was
the sport of fear, folly, and superstition.68
The implication that we have eradicated fear, folly, and super-
stition was unfortunately premature. So too was Strachey’s belief
that scientific methods have taken their proper place in our system.
It is probably an exaggeration – but one nearer, I believe, to the
truth than Strachey’s view –
that all we have done is to dress
our fear, folly, and superstition in respectable garb.

Indeed, in some respects our present state is worse than our
earlier one. Fear, folly, and superstition were in the past tempered
to some extent by considerations of ethics, justice, and humanity.
The incorrect notion that science consists of measurement and
analysis has led us to downgrade those considerations because
they cannot be measured or precisely analysed. And in so doing
we have lost much. The detailed provisions of the American Law
Institute’s Model Penal Code may be more capable of precise ap-
plication to cases of homicide, but they lack the intuitive under-
standing of human nature displayed by Sir Michael Foster in his
celebrated Discourse on the topic.

67L. Strachey, Elizabeth and Essex, (1928).
60 1bid., chapter 6.

No. 2]

IMPLICATIONS OF SCIENCE FOR THE LAW

We are not behaving scientifically when we eschew consider-
ations of ethics and humanity because they cannot be measured,
or because we have no special expert advice about them to guide us.
We are concerned in the law with the regulation of human behaviour,
and such considerations are an integral and essential part of the
situations that we deal with. To leave them out of account is to adopt
an unscientific approach, for it involves the conscious disregard
of essential data.

I return, then, to my starting point. Science is the dominant
feature of our age, and its methods account for its success and
consequent pride of place. If we wish our legal system to achieve
similar success and public respect, we should adopt its methods
and established teachings wherever we can. Specifically, this would
involve, at the least, a re-organisation of our methods of ascertaining
the facts at issue, a reconsideration of our rules of evidence and
of substantive law to ensure that they are based on a correct
understanding of human nature and human behaviour, a re-evalu-
ation of the role which decisions of the past should be allowed
to play in the formulation of decisions for the present and the
future, and the incorporation into the law of a continuing feedback
system. These steps would doubtless require us first to undertake
an agonising re-appraisal of everything that we are at present doing;
and it is to be expected that the implementation of changes would
require much time, thought, and informed debate.
But surely we ought at least to make a start.