1
Introduction: Deep Thought
Introduction
Deep Thought: a brief history of
thought experiments
On the face of it, thought experiments are a useful way to gain new
knowledge about the world, by means of ‘armchair philosophy’ only.
And, whether they are called thought experiments or not, the approach
has had an important role in not only theoretical philosophy,
but in practical science over the centuries.
The ancient Greeks particularly liked to explore using the technique.
Not that they had no concept of more conventional experimentation
too. Empedocles (495–435 BCE), who wisely divided the
world between two forces, ‘love’ and ‘strife’, also founded one of the
first medical schools, from which source a fragment of writing records
a very practical investigation of the circulatory system. But Heraclitus
(c.500 BCE), who liked to write in riddling epithets like the famous
one about it being impossible to step into the same river twice,
decided that as ‘all is flux’, it is ultimately by the power of the mind,
which can contemplate ‘what is not’, rather than by senses forever
limited to examining merely what is, that the truly important things
can be found.
Ptolemy (87–150 CE), the inspiration of future mathematicians and
geometers, as well as geographers and cosmologists, describes his
homely view of the universe in the first book of the Almagest, and
gives various arguments that sit somewhere between ‘thought experiments’
and real experiments. In particular, Ptolemy argues that
since all bodies fall to the centre of the universe, the Earth must be
fixed there at the centre, otherwise falling objects would not be seen
to drop towards the centre of the Earth. Now his listeners could at
this point have conducted their own ‘real’ experiments but it was
clearly enough to reflect on their own underlying assumptions of
WBAA02 7/7/04, 3:33 PM1
2
Introduction: Deep Thought
reality to agree with Ptolemy. Only the fact that the first assumption
is rather a big (and rather a dubious) one stands between them and
true knowledge, a fact that is certainly a reminder of the dangers
of the thought experiment technique. But then again, it is also a
reminder of the dangers of the ‘thoughtless’ experiment technique
too. It was not the lack of testing that was a problem with Ptolemy’s
theory; it was the assumptions underlying it.
Another of Ptolemy’s experiments is designed to show that the
Earth must not only be at the centre of the universe, but completely
motionless – steady as a rock – too. To do this, Ptolemy asks us to
consider the fact that that if the Earth moved, as some earlier philosophers
had suggested, then certain bizarre consequences would
have to follow. In particular, if the Earth rotated once every 24 hours,
was it not intuitively obvious that an object thrown vertically upwards
would not fall back to the same place, but would fall back
slightly to one side?
Ptolemy’s record is not encouraging, but then his experiments were
not truly thought ones. For Plato, as for Heraclitus, those wishing to
understand phenomena in the natural world should recognize that
experience of events is a poor guide. Plato’s dialogues are littered
(for want of a kinder term) with thought experiments. Alongside
Gyges with his magic ring exploring the nature of morality is the
‘mad friend’ hunting for his knife; elsewhere there is Plato himself
conducting the (less well known) ‘breeding experiment’ in which he
advances the case for eugenics for the good of society; and over
there is the much repeated but little agreed upon metaphor of the
prisoners in the cave, that seems to be telling us something about
the nature of knowledge. Less often appreciated, but still very influential,
is that the entire process of the development of society outlined
in the Republic is actually a carefully crafted thought experiment,
built on the assumption that people will not be content with all the
fruits of nature – but will want to eat meat, leading to a struggle for
land and resources.
But it is in Plato’s account of Socrates leading the slave boy, Meno,
to develop the Pythagorean principle that new knowledge appears to
emerge from introspection in the best manner of the technique. Many
of the Ancients valued such ‘pure’ knowledge, quintessentially mathematical,
more highly than any that relied on actually going out and
looking at real things, and the notion of ‘truths’ waiting to be discovered
by contemplation is appropriately sometimes dubbed ‘Platonist’.
Thought experimenters are his fellow travellers.
Even Aristotle, who like a certain kind of scientist usually maintained
the supremacy of observation, tried one or two thought
experiments. In his Metaphysics (Book VII, iii) for instance, he offers
WBAA02 7/7/04, 3:33 PM2
3
Introduction: Deep Thought
the experiment of two individuals, Plato and Socrates, having their
‘non-essential’ properties stripped away leaving only their ‘essence’.
How many essences are there, he asks? One or two?
But Aristotle’s importance in the history of the technique lies less in
his use of it than in his provision of a wealth of poorly judged and
false beliefs about the physical world. As Bertrand Russell has pointed
out, Aristotle, in spite of his reputation, is full of absurdities. For
instance, Aristotle insists that the blood of females is blacker than
that of males; that the pig is the only animal liable to measles; that
an elephant suffering from insomnia should have its shoulders rubbed
with salt, olive-oil and warm water; and that women have fewer
teeth than men. But there have also been more weighty opinions
about gravity, time and space that subsequent philosophers and
scientists have had to labour mightily to demonstrate the error of.
And often the most compelling refutations have been not empirical,
but conceptual, using the thought experiment technique. (Although
not, admittedly in the matter of numbers of teeth. There is still a
place for observation.)
Medieval philosophers, for example, used the idea of a lance with
a sharpened handle (that is, as well as a sharpened point) to disprove
Aristotle’s theory that things like lances only flew through the
air when released, rather than falling bemusedly straight to the
ground in shocked realization that they were no longer being held,
by virtue of the ‘pressure’ of air rushing in behind them. (The air’s
ability to press on the sharpened handle did not seem to them likely
to affect the lance.) The medievals particularly valued such thought
experiments in their debating technique of ‘challenges’, which used
all kinds of ‘common sense’ experiments to settle disputes. In a
formalized process known as ‘obligationes’, disputants were ‘obliged’
to either assent, dissent or doubt statements, until such time as a
‘contradiction’ was demonstrated in one or the other’s position. A
double-pointed lance, although easy enough to produce to the court,
was not necessary.
But it was the Renaissance that produced the richest crop of thought
experiments, including those of Galileo, Descartes, Newton and
Leibniz. These were thinkers whose interests lay in ‘Natural Philosophy’
and who considered that the best experiments work by
making conscious and obvious what any assumed laws of nature
really are. Descartes used the technique particularly enthusiastically,
offering in his Meditations (1641) the original ‘brain in a vat’ scenario,
along with a ‘possible world’ peopled by automata, another run by
a ‘malicious Demon’ (along with the general philosophical problem
of whether we might all be dreaming) and finally the solitary introspection
in the celebrated Second Meditation. It is there he finds that
WBAA02 7/7/04, 3:33 PM3
4
Introduction: Deep Thought
he cannot even imagine thinking away thinking and so is led to the
conclusion that the only certain thing is thought itself.
Descartes took it for granted that whatever can be imagined is
possible, in some sense. Yet although this might appear to give the
imagination extraordinary power, he also insists that mere mortals
are bound forever to the laws of logic, unable to even imagine a
world in which, say, 2 + 2 did not equal 4. (Although God, Descartes
piously adds, is above these laws.) Thought experimenters who
dare to suppose an illogicality enter dangerous waters in which,
even if they manage to survive, any findings they may eventually
return with are worthless. Yet just what is ‘illogical’? In another
experiment, Descartes says if you remove all the matter from a
chamber, the walls would touch, therefore a vacuum is impossible.
Perhaps then ‘imagining the impossible’ is not always so foolish a
thing.
Hume, like Descartes, considered that ‘conceivability’ equals possibility
and that things which cannot even be imagined definitely could
not be possible:
’Tis an established maxim in metaphysics that whatever the mind clearly
conceives includes the idea of possible existence, or in other words,
that nothing we imagine is absolutely impossible. We can form the
idea of a golden mountain, and from thence conclude that such a
mountain may actually exist. We can form no idea of a mountain
without a valley, and therefore regard it as impossible.
Sometimes (but, as we have already seen, quite erroneously), the
philosophical examination of thought experiments is only traced back
to the Danish scientist, Hans Christian Oersted (1777–1851). Oersted
saw them as not so much concerned with predictions or substituting
for measurement, but as a tool for arriving at a better understanding of
nature. For him the value of the technique lay in first of all supposing
some kind of ‘law of nature’ and then asking the experimenter to
apply the law in a new – perplexing – setting. This was at a time
when German philosophers such as Johann Fichte (who had chosen
for himself the task of finding a transcendental explanation for consciousness)
or Friedrich Schelling (nowadays hailed as the father
of ‘post-metaphysical thinking’) were indulging in ‘speculative philosophy’.
Oersted himself has been said (perhaps like Kant too) to have
been looking for ‘a middle way’ between blind laboratory science
and fruitless metaphysical speculation.
Certainly, in the history of science, the thought experiment has to
be acknowledged as a scientific method in its own right. Galileo did
not actually drop balls off the leaning tower of Pisa – it was a thought
WBAA02 7/7/04, 3:33 PM4
5
Introduction: Deep Thought
experiment. (Despite what some may say today, see experiment G.)
Similarly, Leibniz’s procedure for refuting Descartes’ Law of Collision
does not require the rolling around of variously sized billiard
balls; the thought experiment alone is rightly seen as settling the
matter.
Although it is not perhaps a very interesting experiment, it is a
good example of its kind. Descartes thought that if a smaller object
hit a larger one it would rebound with equal speed, and that when a
larger object collided with a smaller one, then the two would move
off together (in a way that conserves the total quantity of motion).
Leibniz, however, asks us to imagine a series of collisions, in which
one ball starts by being smaller, but the ball it is hitting is shaved
down imperceptibly until the first ball actually becomes slightly the
larger. At this point, according to Descartes, the behaviour of the two
balls radically changes. But it seems ridiculous to suppose that such
a tiny change in the ball’s mass could result in it one minute bouncing
off, and the next propelling the other onward, so Leibniz seems
to show that Descartes is wrong.
And much of modern physics is built not upon measurement but
on thought experimentation. Einstein did not carry out measurements
in a rapidly descending elevator, nor did Schrödinger actually
put his cat into a box with a radioactive rock; all were sufficient in
themselves just as hypotheticals. They are quite possible as practical
exercises, but the point of a thought experiment is that it really
would not help to carry it out: all the information that is necessary is
already there, as it were, in the hidden recesses of consciousness.
And in fact Galileo, Newton, Darwin and Einstein all used them to
great effect to resolve, not just explore, complex issues and scientific
debates. They conjured up scenarios, obliged others to follow the
logic of the tale and ultimately accept their findings. These were
quintessentially experiments that took place truly in the ‘laboratory
of the mind’.
Einstein, a past master, used the technique to imagine what it
would be like to travel at the same speed as a light ray. If you were
to run down a pier, he mused, as a big wave was coming in, then the
watery wave would appear to you as a stationary lump in the water.
What then, for an astronaut racing a light wave – would it too appear
to be stationary (experiment U)? In another thought experiment, a
physicist has been drugged and wakes up in a box being pulled
steadily upwards by a rope. Into this box a beam of light is projected.
The ‘elevator’ as it became known, is designed to demonstrate the
equivalence of constant acceleration and gravitational field effects,
by showing that the light ray will appear to bend in both cases. From
such simple musings would come the special theory of relativity. As
WBAA02 7/7/04, 3:33 PM5
6
Introduction: Deep Thought
Einstein wrote later, ‘from the very beginning it appeared to me
intuitively clear that, judged from the standpoint of such an observer,
everything would have to happen according to the same laws
as for an observer who (relative to the earth) was at rest. For how,
otherwise, should the first observer know, i.e. be able to determine,
that he is in a state of fast uniform motion?’
Some say all this is a little too good to be true. They worry that
whilst the approach seems to offer the advantage, through being
made up as you go along, of allowing extraneous detail and complications
to be removed, it may equally in so doing cease to be
relevant or accurate. As Wittgenstein put it, ‘it is only in normal
cases that the use of a word is clearly prescribed, we know, we are
in no doubt, what to say in this or that case. The more absurd the
case, the more doubtful it becomes.’ If you imagine things differently
from ‘the way they are’, he adds, then ‘you can no longer imagine
the application of certain concepts.’ This sits a little uncomfortably
alongside, of course, his own liberal use of the technique, including
examples such as the one describing a man whose brain is removed
by surgeons, and another where we are asked to imagine a world
in which all human beings look exactly alike (compare with experiment
I), not to forget numerous other, at least ‘quasi’, thought
experiments supposedly highlighting aspects of language. Ones such
as the comparison with the controls of the train engine, or the one
with a map of a street accident, or most elaborately of all, the ‘beetle’
that everyone carries furtively around in a small box (see experiment
W). But then Wittgenstein (or at least the ‘later Wittgenstein’
as aficionados put it) did believe that language is best undersood as
a series of pictures, and his thought experiments are also, in their
way, only the logical manifestation of that approach.
In any case, the accusation of ‘abnormality’ also sits rather uncomfortably
alongside the rich history of mathematics, where impossible
entities appear without so much as a raised eyebrow – be they
dimensionless points, perfect circles, negative and irrational numbers
or whatever. Mathematics is after all one of the main sources
of thought experiments in a tradition stretching from the ancient
philosophers up to the present day. Those proposed by Bertrand
Russell, Gottlob Frege and others to resolve the so-called set paradox
(a debate in which Wittgenstein himself was closely involved)
are rightly recognized as central to both the philosophical and the
mathematical debate. And here, in the concluding ‘How to’ guide,
we consider mathematically (but only in the most elementary mathematics!)
a kind of meta-thought experiment designed to show how
the technique might quite legitimately conjure up ‘new’ information
from old facts and established assertions.
WBAA02 7/7/04, 3:33 PM6
7
Introduction: Deep Thought
In fact, mathematics and physics operate with different rules, and
should be kept apart to some extent. Physics is empirical, based
on measurement, but mathematics is based on ‘axioms’ that are
assumed at the outset. Having said that, nowadays, physicists, if not
philosophers, see even mathematical knowledge as provisional and
flawed. Indeed, Einstein once wrote: ‘as far as the propositions of
mathematics refer to reality, they are not certain; and as far as they
are certain, they do not refer to reality.’
But for some philosophers, uncertainty and provisional knowledge
are always going to be a bad thing. Many would-be moral scientists,
such as Utilitarians equipped with simple rules based on the maximization
of happiness principle, worry that thought experimenters
parade ‘a succession of bizarre cases’ which actually warp judgement.
Far from investigating moral intuitions, as they may imagine they
are doing, they are replacing complexities with simplicities, while
always supposing that doing so makes no difference. One critic,
Alisdair McIntyre, has also objected that ethical thought experiments
are ‘ahistorical’ in that they are detached from their original origins
and debates. Others have warned that the thought experiments become
like a stage play in which we are asked to become actors
trapped into endlessly repeating the same scenario, proving nothing.
Marxists disparage ‘soft escapism’, and insist that philosophers stick
to practical issues.
Still others object to the reliance on intuitions, and return to the
age old concern of the relationship of ‘conceivability’ to ‘possibility’,
the debate that used to feature so prominently in attempts to work
out whether God really did exist. But the same concerns apply to
more immediate questions raised by thought experimenters. For
example, on a medical theme, at what point would a person cease to
be alive if body parts were progressively taken away? Such experiments
appear entirely conceivable, but perhaps it is an illusion of
conceivability, an unwise and fruitless adventure into hypothesis.
This is certainly what Ernst Mach, who used the technique himself,
meant when he complained about Newton’s famous ‘Bucket’ experiment
(experiment N), a generally mundane account of a bucket on
a rope in which, nonetheless, Newton surreptitiously imagines
the whole of the universe away. As Mach commented drily: ‘When
experimenting in thought, it is permissible to modify unimportant
circumstances in order to bring out new features in a given case, but
it is not to be antecedently assumed that the universe is without
influence on the phenomena in question.’
The most implacable enemy of the technique, however, has been
a certain kind of traditional ‘analytic’ philosopher, apparently concerned
that he or she is being encouraged to infer conclusions from
WBAA02 7/7/04, 3:33 PM7
8
Introduction: Deep Thought
‘intuitive reactions’ rather than by a sound process of rational deduction.
(Although funnily enough, another philosopher, Richard Rorty,
has said that thought experiments are ‘circular’ because our beliefs
determine what happens in them. Perhaps the concern is that the
technique is stepping on the toes of analysis, as there the aim is to
start by assuming something, examine it a bit, and then conclude
with your initial assumption, having apparently forgotten that that is
what it was. No new information can ever be obtained by ‘analysis’,
as any logical pedant can tell you.)
Certainly it seems these days that many contemporary philosophers’
main interest is in debunking the technique. Thus, a paper
in the journal Ethics by Tamara Horowitz spills much ink in the
cause of denying the hapless Warren Quinn the right to use the
technique to draw conclusions about ethical values. His experiments
revolved around some ‘rescue dilemmas’, the infamous imaginary
underground trolleys packed with different assortments of people
heading forever to various sorts of disaster. In Quinn’s examples, the
numbers of people being rescued (or run over) stay the same, but
the circumstances and indeed the language used to describe them
are varied. Horowitz points out that people are inclined to forgive
unfortunate effects if they are described as incidental whilst condemning
those in which the consequences are described more
explicitly.
In some respects, this is only common sense, but to Tamara
Horowitz it shows rather that responses to thought experiments in
general, and ethical dilemmas in particular, will be influenced
by the wording, or what is sometimes termed the ‘framing’, of the
question. This at least is something we must be aware of both in
considering other people’s, and even more so in designing our own,
thought experiments, and is a point we shall return to in the ‘How to’
section at the end of this book.
On the other hand, there is nothing in the technique that says the
experimenter is not allowed to change the wording if in fact it
is shown to be skewing the results. Moreover, in a stout (if still also
somewhat wishy-washy) defence of the technique, in the Journal
of the Theory of Social Behaviour, Francis Roberts argues that at
least thought experiments allow for investigations to be carried out
without ‘disturbing the environment’ in the process, while elsewhere
Jonathan Dancy, in an account of ‘The Role of Imaginary Cases in
Ethics’, says that they can be just as good as real examples, particularly
when fleshed out with details, even if they suffer from ‘a certain
indeterminacy’ not to be found in reality.
Such as in Franz Kafka’s story, Metamorphosis (1915), which describes
what it would be like to find you had woken up and your
WBAA02 7/7/04, 3:33 PM8
9
Introduction: Deep Thought
body had turned into that of a giant insect. If you believe in reincarnation,
the possibility is quite real, of course, and in the story there
is also a ready made ‘mind-transfer’ built in, of the kind that has
spawned so much vigorous recent philosophical debate. Still, even
ridiculous stories may tell us more about both our intuitions and
assumptions, be they methodological or ethical, than those handcuffed
to reality.
That was certainly the feeling of Charlotte Perkins Gilman, whose
literary utopia Herland (published coincidentally in the same year as
Metamorphosis) described how three male explorers stumble upon
an all-female society in a distant part of the earth. Many generations
earlier, this commune had become separated from the rest of the
human race, with the men dying off. The society had evolved in a
distinctively feminist way, organizing itself around raising children
and living in harmony with its surroundings.
Originally published at the time of the Suffragettes in England and
campaigns in Europe and America for women’s equality, Charlotte
Gilman’s story is a vehicle for her view of male/female roles and
behaviour, of motherhood, individuality, sexuality, and other topics
as well. But more recent feminists have had their doubts too: that
the approach utilizes a ‘restrictive male form of thinking’ that should
instead be trained into a more holistic, inclusive and co-operative
mode. Or that it elevates ‘abstract principles over contextual solutions’,
as Carol Gilligan has put it.
This is a little unfair. One of the most celebrated of recent ethical
thought experiments comes courtesy of Bernard Williams, in which
he imagines a man, Jim, who arrives in the town square of a South
American republic, to find 20 locals there, firmly tied up, and standing
over them the Captain of the local militia. To Jim’s alarm he
declares that he has just quashed their rebellion and is going to
shoot them. Unless, that is, perhaps Jim, as a distinguished visitor,
would like to shoot the first one – in which case the rest can all go
free.
This experiment is intended to challenge precisely that sort of
‘elevation of abstract reason’ (in this case, utilitarianism) and kind
of ‘contextless’ thinking. In running the experiment we begin to
doubt whether such matters can be swiftly resolved by calculating
the ‘amount of happiness’ that results, and overcoming scruples in
order to save the 19 unfortunates. We have also to consider whether,
much as we might like to help, we would be able to justify to ourselves
the sacrificing of the first one.
But in any case, other feminists recently have not been above
using the approach themselves – and with some celebrity. Sissela
Bok, in her book Secrets: On the ethics of concealment and revelation,
WBAA02 7/7/04, 3:33 PM9
10
Introduction: Deep Thought
used a thought experiment with four imaginary societies in order to
examine the issues of secrecy and confidentiality in activities such
as psychotherapy and spying. In experiment V we will see Judith
Jarvis Thompson invent an ‘unconscious violinist’ to test assumptions
over the ethics of abortion, this one of the most successful and
fruitful of contemporary experiments. To join in the debate, Mary
Anne Warren, for example, unplugged the violinist’s unfortunate
neighbour and asked for a volunteer instead, while Roger Wertheimer
came up with a scenario asking what would people think if wombs
were transparent.
And like the feminists, whatever their personal doubts, even the
most analytical of philosophers have also been unable to keep off the
thought experiments. The great German logician, Gottlob Frege, overcame
any qualms to conjure up the possibility of a ‘rational tribe’
who had an ‘alternative logic’ before concluding, sadly, that such a
tribe could no longer be counted as ‘rational’. (For Frege, like his
countryman Immanuel Kant, it was important to demonstrate that a
rule of logic is binding on all of us, from whichever tribe.)
Amongst other more recent efforts are Peter Strawson’s imaginary
‘world of sounds’ in which ‘position’ is determined in some complicated
way by the gradually changing pitch of what is called the
‘master-sound’ (this is discussed further in the ‘How to’ guide) and
Martin Hollis’s strange village (strange, too, in its resemblance to
one imagined earlier by Gottlob Frege) in which anthropologists
struggle to translate their language for fear that ‘the natives’ may
use a different kind of logic to our own. Then there’s Brian Ellis who
wondered whether if the universe had just one thing in it, which he
calls ‘e’, perhaps hoping to nominate himself (Essence of Ellis), that
object could still have ‘quantitative’ properties? How big, for example,
can Essence of Ellis be considered to be, when there is nothing to
compare it with? No measures or rulers, no trees, no nothing.
Or there is Anthony Quinton’s (perhaps rather feeble) effort to
imagine the mind of Winston Churchill in the body of a 6-year-old
girl (with a view to countering the notion that certain character
traits require certain physical prerequisites) and of course John
Searle’s Chinese Room. This last has become quite a regular both
in mass media discussions and on the lecture circuit, and I myself
have not shied away from offering a ‘souped-up’ version here as the
second part of experiment R.
John Searle originally introduced his ‘Chinese Room’ thought
experiment (then titled, ‘Minds, Brains and Programs’) with the explanation
that ‘one way to test any theory of mind is to ask oneself
what it would be like if my mind actually worked on the principle
that the theory says all minds work.’ He hoped his experiment would
WBAA02 7/7/04, 3:33 PM10
11
Introduction: Deep Thought
persuade people to the opinion that theories of Artificial Intelligence
that award computers human-like thinking skills are ‘bunkum’ or, at
the very least, ‘implausible’. Other analytic philosophers, however,
while sharing this aim by and large, cannot accept the proof offered
by a thought experiment anyway. One such, Daniel Dennett, objects
that as the Chinese Room scenario is not an argument, it therefore
cannot be ‘sound’. Unsound, that is, in the sense that even if all the
assumptions made in it were acceptable, no one would be logically
bound to accept its conclusion. Of course, this cuts both ways: even
if you found Searle’s conclusion quite ridiculous, there is no logical
reason in that to suspect either the assumptions or the method
followed.
But while writing a special introduction to a new Journal of
Artificial Life, Professor Dennett also observes that philosophers have
always ‘trafficked’ (as he unkindly puts it) in thought experiments.
These techniques, he noted, are ‘notoriously inconclusive’. ‘What
“stands to reason” or is “obvious” in various complex scenarios is quite
often more an artefact of the bias and limitations of philosopher’s
imagination than the dictate of genuine logical insight’ he says. Yet
even so, there is hope for the technique. Searle may turn on his
swivel chair, but for Dennett at least the computer can make the
philosophers’ thought experiment worthwhile. By modelling hypotheses
on computers Dennett sees the whole exercise as somehow
validated. As he concludes: ‘Philosophers who see this opportunity
will want to leap into the field, at whatever level of abstraction suits
their interests, and gird their conceptual loins with the simulational
virtuosity of computers.’
Of course this is dreamy nonsense. Philosophers should leave
computers, like video recorders and photocopiers, well alone. It is
only worth mentioning here to show that even the most hard-nosed
‘analytical’ philosophers actually believe in the value of hypothesis
making and testing – which is strictly speaking that most evil form of
illogicality, inductive thinking. The thinking, some readers may recall,
that led Bertrand Russell’s unfortunate chicken to waddle down
out of the coop expecting a handful of tasty grain on the day that the
farmer was planning a special dinner . . .
Much of today’s debate over the validity of thought experiments
centres on this issue. The distinction is between a technique that can
provide new knowledge and one that can only present old knowledge
a new way. On the one side philosophers such as James Brown
say thought experiments provide what they like to call a priori knowledge
of natural phenomena, such as the mathematical entities
or ‘laws of nature’ that mathematicians and physicists wrestle with,
and which may or may not be ‘out there’ in the world of the Forms;
WBAA02 7/7/04, 3:33 PM11
12
Introduction: Deep Thought
while on the other, philosophers like John Norton stoutly maintain
any knowledge obtained from them is not new, not ‘discovered’, but
merely disentangled from where it is already lurking (in the disorganization
of the imperfectly logical mind).
In fact it was John Norton who once grandly defined thought
experiments as arguments which:
1 posit hypothetical or counterfactual states of affairs and
2 invoke particulars irrelevant to the generality of the
conclusion.
But grand though it sounds, to say that a thought experiment is
‘hypothetical’ is to say precisely nothing, whilst to add ‘or counterfactual’
is to play at words. In allowing something to be either counterfactual
or factual we are not venturing very much. In fact, some
thought experiments are counterfactual, but many others seek on
the contrary to demonstrate some facet of reality through all the
elements being entirely possible, if not necessarily plausible. This
leaves only the observation that thought experiments include irrelevant
details as the definition which seems (in a possibly interesting
way) to be exhibiting precisely the fault that it claims of the thought
experimenter.
As to that, consider Galileo’s Ship argument (experiment S) with
its cutesy details such as the fish that swim towards the front of their
bowl or the butterflies that continue their flights indifferently towards
every side. It is guilty on all counts. Yet it is also the basis for
much of modern physics, and created a world in which dogmatic
assertion began to weaken. Up until then, as a result of such arguments,
the geocentric system was forcibly thrust upon all philosophers
and scientists by theologians relying not only on the guidance
of divine texts but the apparent certainties offered by ‘science’. Yet,
for Galileo, the approach of the thought experimenter offered more
certainty and greater validity than any number of measurements
or predictions (leaving well alone the untouchable authority of the
scriptures).
And so, if even today those who follow in Aristotle’s footsteps are
baffled by the whole approach, it is perhaps only the more reason to
revisit the great discoveries and debates of thousands of years of
experimentation in the ‘laboratory of the mind’.
WBAA02 7/7/04, 3:33 PM12