Ernest Nagel on the Philosophy of Science (1954)

The subject that I'd like to talk about

this evening

is a rather vast one.

It is not unified by a common set of

principles

or by what one might perhaps even call a

common subject matter. It is not easy to

define it the way one could for example

define formal logic

or perhaps other special areas of

philosophical inquiry.

The nearest I think I could uh come to

giving a general formula to cover the

subject is that the philosophy of

science is a critical commentary

on science.

The critical commentary itself

might take different directions

depending upon

what the special interest of the

investigator happens to be.

Uh perhaps uh one way of uh introducing

the

uh character of inquiry into science is

to recall some

important historical examples.

uh in which uh science and commentary on

science has played an influential role

in directing men's thoughts one way

rather than another. The two examples

I'd like to mention very briefly

is first of all the very important

Capernac Galilean Newtonian revolution

which brought about a fundamental change

in men's outlook

and which required on the part of

reflective thinkers

an effort to adjust old ideas to the new

ones.

an attempt to reinterpret old values

in terms of most recent discoveries

in one particular area of investigation.

An attempt also to

reconsider

the objectives,

the values that men possessed

in so far as those values and those

objectives depended upon having some

clear conception as to the kind of world

they inhabited.

And so if one examines the literature of

this period, one finds not only simple

expositions

of the technical content of physics and

astronomy,

but attempts to show what the relevance

of

Newtonian ideas, for example, were for

problems of ethics, for problems of

politics.

Attempts were made, for example, to show

that a desirable form of political

organization

ought to take its point of departure

from

the ideas that Newton had developed. And

since it was a period when monarchical

systems

were very much

in favor, attempts were made to use

Newtonian physics as a support for a

monarchical system of government.

Now this is one illustration of one kind

of interest that people do have in

science in wishing to comment on it in

trying to bring it close to men's bosoms

where the attempt is not simply to

expound the content but to reinterpret

it in such a way that it appear that the

conclusions of science appear to have a

relevance for problems of value for

problems of government.

The second example I want very briefly

to mention

is the

change in outlook that took place in the

in the 19th century as a consequence of

Darwin's ideas on the origin of species.

This too shook men's convictions earlier

convictions about the kind of world that

we're living in. not so much the

physical world as the biological world

they were living in. And here again and

in this case I suppose many of us

perhaps may even remember uh the

overtones of the discussions of

Darwinian theory which uh were still

widely current in the first decade of

the present century particularly in this

country where attempts were made to show

the bearing of biological evolution on

problems of ethics on problems of

treatment of various races of mankind on

problems of economics.

Furthermore, in the case of

Darwinian evolution, philosophers too

believe that here they had at their

disposal

a set of ideas in terms of which the

character of human knowledge itself

could be reinterpreted.

so that the mind came to be conceived

not as a passive beholder of a grand

spectacle but simply as an instrument in

advancing the fortunes

of the human species.

Now all these comments on technical

developments

uh have continued to influence

uh our own thought because we inherit

these distinctions and these criticisms

that uh earlier philosophers

uh have offered uh on the detailed

materials that specific empirical

investigations

bring forth.

Now it so happens that uh in the present

century

despite the prophecy that a good number

of people had made

that

physics had seen its best days that the

great achievements in physics had

already taken place

uh in the 18th and 19th centuries that

the 20th century would be the period

during which biology,

psychology

and perhaps the social science es uh

would be making the most uh significant

advances. These prophecies have turned

out

not to be accurate ones. That in point

of fact, the supposition that physics

had already said the last word on all

the fundamental questions about the

character of the universe we're living

in at this conception has been belied as

everyone knows by the developments in

the last 30 40 years.

The philosophy of science usually pays

attention to those sciences

which uh are at the forefront of

research and which introduce new ideas

and seem to involve some sort of a

revolution in the basic conceptions and

attitudes that men hold.

And so if one looks upon examines the

uh developments in philosophy which

bears upon the sciences I think it's

safe to say that in the last 40 50 years

most of the attention has been paid to

developments in the physical sciences

and only incidentally

uh to the sciences which deal with

biological phenomena or which deal with

specifically with human affairs. is

the impact of physics of course has been

very very great upon our own age for

several reasons. One very obvious

reason.

If anybody ever had any doubt that uh

science in particular physical science

is capable of baking bread,

certainly

uh the events of the past 10 or 20 years

uh must have certainly settled those

doubts. that there has been an obviously

increased practical power which science

has yielded and the prophecy that

Francis Bacon made in the 17th century

that the new science would really give

men mastery over nature has certainly

been realized beyond his wildest dreams.

So there has been a social impact of

science as an attempt to readjust

customary ways of living in the light of

new instrumentalities.

This has raised fundamental problems in

government, fundamental problems in

reorganizing our moral attitudes.

There has been another impact that I

think science has made upon reflective

men. Namely,

the impact that has arisen from the fact

that as the sciences have

advanced and developed,

their content

appears to be much more remote from the

familiar things of everyday living than

the older than than the older physics

has been.

There's there's a kind of a paradox

about this. Uh many people are con have

have developed the attitude or have have

developed the sentiment at any rate that

uh though our practical mastery over

nature has increased.

Nature progressively seems less

intelligible

and has become less intelligible because

current theories in the physical

sciences are so abstract, so remote from

the affairs of everyday living that uh

uh any attempt to interpret them in such

a way that the plain man in the street

who hasn't the technical knowledge to

follow the details of a technical

argument

uh seems seems simply bewildered by the

paradoxical announcements that

scientists make about the executive

order of nature.

And so there has been a a a both a a

challenge to our traditional beliefs as

well as a challenge to

conceptions as to what is the nature of

scientific knowledge.

And it is really about the second of

these that I want to spend most of my

time this evening.

But before I say something more specific

about uh the issues that have arisen

because of the progressively greater

abstractness of physical theory. I'd

like as a as a preliminary to map out in

a way four major areas

uh into which the philosophy of science

can conveniently be divided.

Uh so at any rate you will have a a a uh

a bird's eyee view of the various sorts

of things that people do discuss when

they engage in this activity of

philosophical commentary upon

contemporary science.

The first area perhaps for the lack of a

better name I'd like to call after a

phrase that uh the late uh uh Alfred

North Whitehead used is a critique of

abstractions

and which is intimately related with the

point I had just made.

What we try to do in the various

sciences with a greater or less degree

of success is to offer systematic

explanations for the phenomena in our

environment.

The sciences seek to explain and they

seek to explain in such a way that they

don't offer a special explanation for

each individual item. But the

explanations are intended to be formed

in such a way that a single set of

assumption

should cover as wide a territory as

possible

and so theories are formulated with the

help of which this can be done. Now I do

not pretend of course that in all areas

of investigation the same degree of

comprehensiveness is achieved.

The physical sciences certainly

represent the most successful

uh

uh achievements in in in this direction

in the social sciences despite centuries

of effort. I suppose it's only fair to

say that we are only at the beginning of

such explanations.

Uh so though I recognize that different

sciences exhibit this character of

systematic explanation with different

degrees of completeness, yet I think it

is fair to characterize

all scientific undertaking as an attempt

to find systematically explanations.

I do not mean to say, of course, that

there may not be other objectives which

lead men to science. for example, the

obvious objective of trying to obtain

the practical mastery over various parts

of nature.

However, I do not think that this

objective is essentially different from

the objective of seeking systematic

explanation because the sole way that

you can obtain practical mastery

uh is by formulating comprehensive

theories

which enable you not only to order the

facts that you already have but to

foretell to forecast events that have

not yet happened.

So whether you put the stress upon

explanation or whether you put the

stress upon prediction

is simply a matter of which side of a

coin you're looking on. I mean these are

seem to seem to me to be simply obvious

sides of the same fact. Now the point I

want to make here is when I say well now

this is one area of investigation is

that in the attempt I to uh develop such

comprehensive theories it's inevitable

that one must employ ideas that are

highly abstract that are not related in

an obvious way to the familiar things in

experience. And the problem then becomes

a problem with which professional

scientists on the whole are not terribly

concerned in an explicit way, but which

has always challenged the ingenuity of

philosophers to see in what way a

connection can be established between

such abstract notions as space having

curvature

or abstract notions such as those which

involve references to

uh submicroscopic particles and the

familiar things that we encounter in

everyday living. So what I've called the

critique of abstraction is a deliberate

attempt to develop techniques and to

apply them to the concrete materials of

the sciences with a view to showing

what is the bridge by means of which one

passes from things that seem to be

utterly remote

from matters that are familiar to us to

these highly abstract recondite

sometimes paradoxical notions that the

theoretical scientist employ.

Now, if I may pass very quickly onto

this the second area,

the second area I I'd like to again very

briefly refer to as an area which is

concerned with the grounds of certitude

in the sciences.

A contrast is sometimes made between

scientific knowledge

and the kind of knowledge that we

perhaps acquire without benefit of

systematic inquiry by saying well we

obtain a greater measure of certitude in

the sciences than we do when we uh

simply experiment perhaps in a uh trial

or error uh uh manner without taking too

much care about the way we make

observations.

without using complicated schemes of

measurement.

Whether this is uh an accurate

description of the difference between

what one might perhaps call common sense

knowledge and scientific knowledge may

be a matter of dispute.

But there is no doubt whatever that very

important questions arise concerning

the status the intellectual status of

the basic premises the basic principles

in terms of which the scientists offer

explanations.

The question arises why should we credit

them? What is the ground for our

intellectual certitude?

Why should we credit current physical

cosmological theories rather than the

old myths that our ancestors held? What

is the basis for this? In what way is

the evidence marshaled to support these

large claims? Large claims which cannot

be established by simple observation, by

simple looking. When it is said for

example that in the sciences all that we

can hope to achieve are statements which

though we may not know that they are

true at least we know that they have a

high degree of probability. What

precisely is meant by saying that these

statements are highly probable?

Wendy said that we use methods of

induction in establishing the

conclusions of science. What precisely

is meant by saying that we use an

inductive method and how do we justify

the procedures that are called the

inductive procedures? These are all

questions that fall into the second area

that I have here briefly indicated.

I come to my third

uh division and then perhaps uh when I

complete this I might have an

opportunity to say something about

typical figures that fall in who have

been writing in the last 40 50 years who

fall under each of these each of these

divisions. The third the third division

uh of what I'm calling the philosophy of

science is concerned with using the

conclusions of the sciences

to build up a systematic world picture

and attempt to integrate the various

special disciplines

in order that we might have a total view

a total system of the universe.

Now this has been attempted in a number

of ways.

One way that has attracted a great deal

of attention

oh perhaps 15 20 years ago was to try to

formulate an evolutionary schema.

An an evolutionary schema which is

intimately associated with ideas or at

least with with with words with which

many of us are very familiar. a schema

of emergence, schema of historical

sequences.

So that the various items one finds in a

universe for example the level of purely

physical things, the level of purely

biological things, the level of

organization which in includes uh uh

sensitive or or psychological uh

capacities, then the level of the

intellect, then the level of the spirit.

Attempts have been made to show that the

whole universe could be interpreted as

indicating a progressive emergence of

so-called higher layers from the lower

ones.

Now, a task of this kind obviously is a

very difficult one because it involves

uh a really an almost an expert control

over the details of scientific

conclusions.

And although perhaps 2,000 years ago it

was possible for one man to be master of

all the sciences, today it is hardly

possible for a single man to be

acquainted with all that's going on in

one particular area.

And so evolutionary schemes of this kind

have on the whole been judged to be at

best speculative, to be insubstantial

and not very well supported by the

facts. There have been other attempts to

bring about a systematic worldview on

the basis of the conclusions of the

sciences. This has been done by trying

to find some very basic characteristics

that all existence exhibits.

uh if might use uh a technical jargon

for a moment. Uh there's been attempt on

the part of a very large number of

philosophers

uh to formulate a set of categories. A

set of basic distinctions

which are of such a kind that whatever

exists

will exhibit these categories will

exhibit these basic distinctions. And

then the task of the philosopher is to

systematize these categories in such a

way that whatever happens in nature

whether it's physical or biological or

social nature could always be offered as

an illustration of some combination of

these categories.

Uh attempts of this sort have been made

repeatedly by very distinguished

figures.

uh I shall want to say something very

briefly about the success of such

undertakings.

But

uh the development of the philosophy of

science in this direction has occupied

certainly uh the attention of a

considerable proportion of professional

and perhaps even non-professional

thinkers who take their point of

departure from the conclusion of the

sciences. Then very briefly I want to

say something about this fourth division

which will complete this preliminary map

where one is interested in the sciences

not from the point of view of analyzing

the meaning of the terms. Not from the

point of view of

trying to discover the basis for the

kind of assertitude which the scientists

appear to possess. Not from the point of

view of building upon the conclusion of

the sciences in order to get a more

inclusive system. but from the point of

view of trying to evaluate the cultural

basis and the cultural significance of

the sciences.

So inquiries of this kind will include

for example considerations of the

stimuli to the development of science

that is how does science arise in

society?

What sort of a problems are science is

concerned with? What is the what is the

social basis for an interest in one line

of development rather than another? That

is there's a concern simply with the

genesis of science

with uh a consideration of the relation

of theoretical research for example to

certain needs that society at a given

time may have. On the other hand,

there's also the obverse side of this

inquiry. Yes, the problems might be

suggested by the needs of the society at

a given time. Supposing that you take a

view of that sort. Nevertheless, also

there's a sort of a reflex of science

upon society. That is something happens

to society as a consequence as a

consequence of the advances in the

sciences. So, not only is there a

concern with the genesis of science but

also concern with the impact of science

upon society.

And uh then uh if you pursue this matter

of the impact of science of society,

there have been deliberate attempts just

as I've mentioned a little while ago in

connection with the development of the

Newtonian Darwinian uh systems to to

show that

the discoveries of present day sciences

have some very definite implications for

certain moral, political, religious

views that individuals hold. So for

example, I shall have occasion to

elaborate a little bit more fully in a

moment. Many people believe that

discoveries that are made about

subatomic particles such as electrons

have some bearing upon issues in ethics

in particular the issues that are

associated with this age-old controversy

about the freedom of the will. that is

this area of of interest that I'm now

briefly sketching the cultural basis and

the cultural import of science attempts

among other things to ascertain to what

extent if to any extent the special

findings of the sciences have

implications

for the moral life of man and for the

values that men uh decide to cultiv

cultivate.

Now, so much for this preliminary map.

If there were time, I would uh uh like

to illustrate in great deal detail work

that has been done under each of these

ma main areas in the last 40 or 50

years. I can't do this systematically.

uh what I would like to do is uh mention

a few influential movements

and then whatever time I have left I

would like to uh give some examples of

the kind of work that has been done in

the philosophy of science that falls

really under the first head that I have

mentioned namely that part of philosophy

of science which is concerned uh with a

critique of abstraction

Let me however say something about some

recent trends and let me introduce this

uh the series of thumbnail sketches

by making one very general remark.

If one uh examines the conceptions that

men have held about the powers of human

reason from the time of Aristotle

to perhaps the most recent commentary on

developments in modern science. I think

one comes away with the strong

conviction that the claims that

scientists today make

about the scope of reason are much more

moderate than those which were made by

our ancestor 2,000 years ago.

Let me uh give some point to what I'm

saying by reminding you of uh something

which doubtless you've heard uh uh on

some occasion that if you go to the

writings of let's say Aristotle who

really set the fashion as to the nature

of scientific thought for some 1500

perhaps 1800 years

and if you examine what he had to say

about the nature of science science

you'll find roughly he had some he had

the following to say he said we have to

distinguish between

knowledge of a fact and knowledge of a

reason fact.

Knowledge of a fact would be for example

simply a knowledge that well ice floats

on water. You discover that ice floats

on water. You still haven't got

scientific knowledge. You have

scientific knowledge only when you have

shown that this fact that you have

perhaps established by observation

can be shown to be necessary

that things could not be otherwise than

the way that you find them. Now how do

you show that the facts that you

discover are necessary? Well, according

to Aristotle, you have demonstrative

knowledge when you show that this fact

that ice floats on water, for example,

is a logical consequence of something

else. That is, when you give a

demonstration of it, just as you have a

reasoned knowledge of the fact that the

base angles of an isosles triangle are

equal when you show that it's a

consequence of a certain set of axioms.

So Aristotle said you have reasoned

knowledge of the world about you when

you show that the facts that you

discover are

demonstrable consequences of some

fundamental set of axioms. Now, and this

is the the crucial point. If you asked

Aristotle, now how about these axioms,

how do you establish them?

After all, if you're going to prove, you

have to have some premises from which

you are going to make the derivations.

You can't prove everything because every

proof requires some starting point. If

you deny this then Aristotle said you

would have to go on proving things uh

one thing after another and you would be

led to an infinite regress. In

consequence you will never establish

anything. So you must start out with

some principles which are not proved.

How do you know those? Aristotle's view

was you know those simply because the

mind has a capacity of grasping some

truth as being necessarily so. Just as

it used to be held that the axioms of

geometry were self-evident, so Aristotle

maintained that the fundamental

principles of a science had to be

self-evident truths. That the

fundamental principles of a science were

better known were more familiar

were uh less dubitable than any

conclusion that you derived from them.

And then if you asked why do you believe

that ice floats on water? The ultimate

answer is not because you have seen it

but because you have been able to show

that this fact is a demon conseident.

What would some such principles be?

Well, for example, that water expands

when it is frozen.

Because if you assume the principle that

water expands when it's frozen, if you

also assume another principle that when

you put a body into a liquid, the liquid

supports or or exerts a force upon the

the body uh by an amount which is equal

to the weight of the displaced liquid.

If you assume these two principles, you

can prove that ice will float down

water. The details are a little bit

complicated and there's no point in

going through them. But Aristotle

believed that ultimately you came to

some principles which had to be

self-evident. Now the point I want to

make is this. This was not a view that

Aristotle held and nobody else. This was

a view which continued to be held down

to very recent times. It was a view that

was held by Decart who formulated the

philosophy of the new science of the

period. that is who held who formulated

the philosophy of the 17th 18th century

uh pioneers of modern science. It was a

view which continued to be held even

down to the latter part of the 19th

century. Now if you see what scientists

today tell you about their first

principles, I think you are all

impressed by the fact that they would

make no such claim at all. that this

claim for infallible certain necessary

knowledge is not something that anybody

can claim to have with any warrant in

any of the empirical sciences. You might

claim it in mathematics, you might claim

it in logic, but you cannot claim it in

physics or in biology or in psychology

or in any other uh area which deals with

uh matters that fall within observation.

Now I make this preliminary remark

because I think uh the the tendencies

that I want to enumerate are really all

tendencies which enforce the conclusion

that this ancient rationalistic

conception as to the powers of reason

that that the re that that the mind the

human mind is capable of grasping once

for all the ultimate structure of

things.

uh that this is not something that

corresponds to the character of modern

science that if you what you mean by

knowledge is the sort of a thing that

Aristotle said a science must give you

then modern science does not give you

knowledge. I emphasize this point also

because

there have been a great number of people

who have become terribly skeptical about

the kind of a knowledge that science

provides.

Now I suspect that the skepticism which

a great many people have developed is a

consequence of their supposing that the

sole kind of knowledge that is worth

having is the kind of knowledge that

Aristotle mentioned and that whole line

of successive Aristotle regarded as

formulating the ultimate ideal of

science. I think it's important to

recognize that

you are skeptical very often simply

because you are employing standards

which are really not appropriate for a

given material. That is if you are

skeptical about the ability of modern

physics or of modern biology to give you

reliable knowledge,

you are usually skeptical because you

think it does not give you the kind of a

knowledge you can say well now here I

can bank upon this in such a way that I

will have never any occasion in the

future to revise my judgment about this.

And I think if you recognize at the very

outset that this is not a legitimate

ideal to pursue in the light of the way

in which we do obtain knowledge,

then we can really evaluate how

irrelevant the skepticism which a great

many people including distinguished

scientists themselves have expressed

about the capacity of various special

disciplines to give us sound knowledge

of the world that we inhabit. Now then

some very brief comments on recent

tendencies. There's one which in this

country certainly has ex exerted a great

deal of of influence and which is rather

ani an indigenous growth in this country

and only uh gradually has it spread

beyond the confines

of the United States. I have in mind

here of course the attitude or the the

point of view which technically is

called instrumentalism sometimes called

pragmatism and which is associated

uh particularly with the pioneering

writings of a philosopher who has been

neglected for a great number of years

Charles Pur and the writings of uh the

late John Dwey

where the emphasis is has been placed in

their discussion of the character of

science upon the instrumental character

of scientific theory. That is the the

emphasis placed upon this sort of

consideration. What do we want science

for? And the answer is in order to

enable us to pass from one part of

experience to another part of experience

in a continuous and smooth way. So if

you want to know what the function of a

theory is, if you want to know in what

way a theory explains,

what you ought to do is to concern

yourself with the instrumental function

of a theory. Now in the light of this

emphasis upon the instrumental function

of a theory, a great number of specific

questions have been resolved in such a

way that they no longer appear to

represent a an insoluble problem. And I

want to uh uh to talk about this uh in a

in a slightly uh greater detail in in a

few minutes. But rather closely

connected with this emphasis upon

instrumentalism

uh which I say has been an indigenous

growth in this country but which uh has

slowly percolated to other parts of uh

of the world. And so for example at

present it it so happens that a number

of tendencies on the continent in Europe

as well as in England have coalesed with

uh what we're calling instrumentalism

here. So that for example although I'm

do not hap do not I I haven't heard what

uh professor Max black had to say about

uh analytical philosophy at his talk

last week. I'm quite sure that uh if you

examine the relations of analytical

philosophy to American brand of

instrumentalism that even though the

origin of analytic philosophy has

largely been on a continent of Europe or

in England

that there is a kind of an an approach

of one to the other that the emphasis is

towards showing the function of various

parts of language in carrying to a

successful ful termination certain

undertakings that men engage in. Now I

say rather closely connected with with

this uh emphasis upon the instrumental

function of science is the point of view

which uh professor bridgeman at Harvard

has called operationalism.

A point of view which emphasizes

that well whenever you use a term in a

science uh what you have to do is uh uh

to

specify the ways in which this term is

related to the concrete materials of

experience. That is an operational

definition for a concept means

specifying the ways in which you would

apply a term. You have an operational

definition of the term weight when you

specify ways in which you would weigh

objects. You have an operational

definition of a term like electron when

you have specified the ways in which you

presumably would identify electrons.

That is this emphasis upon

operationalism

has had a wide influence. Not so much in

physics

where in a way this emphasis may not

have been needed because the physicists

have long years of of habituation to uh

desirable habits of workmanship.

But in uh the the the beginning sciences

so to speak I mean there's been a great

deal of influence of operation upon

psychology operation upon sociology

where people gradually became to

recognize that they were using language

which did not have any very definite

meaning and bridgeman's injunction that

what you have to do is to specify the

ways in which these expressions are

employed was regarded for a time as

being a solution to all the problems in

these in these areas. And as a matter of

fact, I think the tendency has gone too

far so that a great many uh uh workers

in special areas such as psychology or

sociology feel that anybody who proposes

a theoretical framework in these

disciplines

which employs terms which did not have

an operational definition is talking

nonsense.

I think myself this is a mistake. It's a

mistake because if you turn to the

content of physical science itself, I

mean, if you examine for example a a a

theory like relativity theory or a

theory like the kinetic theory of matter

or a theory like thermodynamics and if

you look at some of the terms that occur

in it, term like electron, a term uh

like uh entropy, a a uh a a term like

instantaneous ous velocity

and you ask yourself now what is the

operational definition of this term?

How would you specify by a laboratory

procedure what you mean by an electron?

If you reflect upon it, I think it

becomes quite clear that you do not give

an operational definition of these

terms. What after all is required is not

that you should give an operational

definition for each single term. What is

required is that your theory be so

formulated that from the assumptions of

the theory you are able to derive other

statements such that eventually you will

get statements whose terms can be

identified

by way of some laboratory observational

operational technique. That is, if I may

repeat this, those who maintain, as so

many current people who are concerned

with problems of language maintain that

a term is meaningless unless you can

give an operational definition.

I think I'm making a mistake because if

they were consistent in their

contention, they would have to throw out

practically all of modern theoretical

physics.

What is required is not so much an

operational definition for every term in

a physical theory or in a biological

theory or in a chemical theory. What is

required is that eventually you can

derive from the fundamental assumptions

of the theory some statements

which can be shown to refer to matters

of observation.

And so we have these highly abstract

theories like the electronic theory of

matter without giving operational

definitions of terms like electrons or

operational definitions of what is meant

by if we use some of these older

conceptions about quantum theory. What

is meant by saying that a an electron

jumps from one orbit to another? There

is no way of identifying operationally

what is meant by a jump. There is no

operational definition for the notion

that an electron revolves in an orbit.

These are purely theoretical ideas. The

important thing is that Boore who

invented this theory was able to show

that from these assumptions,

he was able eventually to derive

statements which bore directly upon such

things as lines in a spectrum of various

elements

as uh defraction patterns which are

formed when you pass light uh through a

grading and the like. That is you have

certain materials of observation. Your

theory must eventually turn uh its

apparatus upon this without requiring

that every term in the theory be tied

down in a very intimate way to the the

uh uh the concrete matters of

observation. And uh this leads me then

simply to mention one other uh one other

tendency

uh rather intimately associated with

these two that I have mentioned

which uh

uh attempts to construe scientific

theories

as being primarily

if not exclusively but at least

primarily

modes of representation

which uh we use in order that we should

be able to bring together into

systematic form a variety of observable

facts. Let me illustrate this a little

bit in in terms of uh an example that I

try to make as as simple as as possible.

First let me take a a an uh a slightly

more recendite example and then uh give

a give a more simple one. uh uh those of

us who read a little bit about the

history of science may recall that when

Faraday

who was primarily an experimentter

didn't have very much training in

mathematics

uh when he was carrying on his

experiments on magnetism and on

electricity

uh he supposed now that uh there are

lines of force which fill the which fill

the medium which fill space. How did he

conceive of these lines of force? Well,

he thought of them really as kind of

tubes and that the intensity of the

force depended upon how thick these

tubes were. But they were, if you

remember some elementary physics, they

sometimes represent as there were rubber

bands. Uh when you stretch these rubber

bands, then they become thin. Other

times they might be thick. and the

strength of the magnetic field or the

electric field then is interpreted in

terms of these lines of force. Now from

the point of view of one type of

analysis what would be said about

Faraday's ideas was well you see you

don't have to take this notion of lines

of force literally you don't have to

suppose now that there really are these

tubes filling all of space that this is

simply an extremely convenient and

fertile way of formulating certain

relations between the behavior of bodies

that you can observe.

Now let me expand upon this point now in

a slightly different context.

Uh take the familiar theory that light

travels in a straight line.

Now let's ask ourselves one very simple

question. Have we ever seen something

that we can call light?

Well, of course our tendency perhaps

would be to say yes, we have seen light.

But uh if you think about this for a

moment, you say, "Well, what you have

seen of course are illuminated objects.

I can see the wall. I can see the sun. I

can see the sun's rays." Meaning by that

that if there are dust particles in the

air, I can see these dust particles

which are illuminated by the sun. But in

the sense in which in elementary physics

we talk about light moving in straight

lines. We haven't seen light moving

is light is not something that we see

moving at all. I mean here we are in

this room we say well now there are

bulbs burning. Uh we don't see those

lamps moving. We don't see anything

moving.

Why is it that we talk this way? Well,

the answer is in terms of one mode of

analysis is this is a very convenient

way of talking because if I talk this

way, I can explain an awful lot of

things. Well, what can I explain? Well,

I can explain such things as that

objects have shadows

and I can calculate the length of the

shadows

on the assumption that light moves in

straight lines. Then if you remember

some very ele elementary ele elementary

ele elementary ele elementary ele

elementary ele elementary ele elementary

ele elementary ele elementary ele

elementary ele elementary ele elementary

ele elementary ele elementary ele

elementary ele elementary ele elementary

ele elementary ele elementary ele

elementary element geometry if you know

the angle at which the the line is

inclined to you and if you can then

measure the height of the object you can

calculate the length of the shadow or

vice versa if you know the angle at

which the line is inclined to you if you

can measure the length of the shadow you

can measure the height of the object. So

if you do not know what the height of a

mountain is or if you do not know what

the height of a tree is but you can

measure the shadow which is cast by the

sun and can measure the angle between

the sun and the uh and the line which

the line formed between the sun and the

top of the tree or the top of the

mountain and the ground then you can

make these calculations.

So to talk this way uh talk about light

moving in a straight line is a

convenient way of making intelligible to

ourselves a great many things and they

become intelligible because you

establish connections between them.

So this emphasis again upon theories

simply as being modes of representation

see is is is really very closely uh

related with this conception that the

function of a theory is an instrumental

one that is has an instrumental function

namely enabling us to bring together

into a systematic

set of uh relations phenomena that

appear to be entire highly desperate.

Now I'd like to uh leave myself uh

perhaps just a few minutes in order to

talk about one uh one particular point

which really falls under the first

heading uh of these four divisions that

I've mentioned

uh

and which uh is really concerned with a

problem of critique of of abstractions.

And I'd like to uh take a few minutes

then uh to discuss what has so been so

so been heatedly discussed in uh in

recent years. The status of certain

familiar principles which apparently

have been challenged by recent

developments in physical theory. Uh

there are a whole set of such principles

which have been challenged. The one that

I propose to take for discussion is the

problem of causality.

And I uh although obviously I can only

scratch the surface and uh must

necessarily be extremely superficial. I

would like to indicate a line of

analysis of this problem and in this way

perhaps help to the extent that I can in

a few minutes uh uh eliminate possible

confusion as to what this problem is.

Now a great many people have been

concerned with the question whether

modern physics has shown us that we are

living in a world in which causal

relations are are are are absent.

Why is this a problem today? Well to

understand why it's a problem I think

one have to look back a little bit about

the history of of of physics itself.

uh in the 17th century

the one branch of physics

that was very welldeveloped was the

science of mechanics.

It was believed by the men of the of the

17th century that the future of physics

was identifiable with the future of

mechanics.

uh men like Huygens,

like Daycart, like Newton believe that

eventually

all parts of nature would be

understandable

on the basis of the fundamental ideas

that the science of mechanics itself

employs. Now, what are the fundamental

ideas? Well, briefly, if we may uh omit

all details, you say, well, mechanics of

course uses the notion of length and

various combinations of an area of

course and volume which are definable in

terms of length. It use the notion of

time. You use the notion of mass.

Well, if you have length and time, then

you can define velocity.

If you have length and time and mass,

you can define force and so on. So there

are a lot of things that you can define

in terms of these fundamental notions of

mass, length and time. Now if you look

at the way in which mechanics works, you

find that you have to do the following.

That in order to be able to predict the

path of a projectile, for example, you

shoot a uh you you you you throw a ball

and if you want to know where the ball

is going to be at some future time, and

this is a problem that can be handled in

terms of mechanics, what must you know?

Well, you must know of course the laws

of mechanics.

You must also know what are the forces

that are acting. And in this case, we

might suppose now that there are the

gravitational force which conforms to

Newton's famous formula that there's a

certain impact that is given to the

bullet. Let's suppose that we all we

know all these things. What else do you

have to know? Well, you have to know two

very important things. where this bullet

was at a certain time

and what its velocity at that time was.

Look, let me repeat this. In order to be

able to predict the behavior of the

bullet in terms of mechanics, you have

to know the position and the velocity of

this body at a certain time. Now,

position and velocity are called in

mechanics

the coordinates of state. The

coordinates of state in mechanics are

position and velocity. Now why do I

stress this? Because if you ask what is

meant by causality in classical

mechanics, the answer is very simple.

Once you know the mechanical state of a

system for one time,

you can in principle predict the

mechanical state of that system at any

other time.

If you know where Mars is at a certain

time and what its velocity is, assuming

now that you are given this other

information that I mentioned, I mean the

laws of mechanics, the forces that are

acting, then in principle you are

capable of making a complete prediction

about the future of this object.