Science: distinctions, aims and criteria

SOURCE: Introductory Readings in the

Philosophy of Science, by Klemke,
Hollinger and Kline (1988)

I. SOME DISTINCTIONS

(1) Pure sciences versus applied sciences.
It is widely held that we must distinguish:
(A) science as a field of knowledge (or set of cognitive disciplines) from
(B) the applications of science.

(A) Among the pure sciences we may distinguish:
(a) the formal sciences, logic and mathematics;
(b) the factual or empirical sciences.
(b1) the natural sciences, which include the physical sciences,
physics, chemistry, and so on, and the life and behavioral
sciences, such as biology and psychology;
(b2) the social sciences, such as sociology and economics.

(B) The applied sciences include the technological sciences - such as:
engineering and aeronautics, medicine, agriculture, and so on.

TWO levels of application among the various sciences
1. the application of the formal sciences to the pure, factual sciences. Since the
factual sciences must have logical form and usually utilize some mathematics,
such application is often held to be essential for the development of the pure
factual sciences.
2. the application of the factual sciences to the applied sciences. Here the
findings of the pure, empirical sciences are applied (in a different sense of
applied) to disciplines which fulfill various social, human purposes, such as
building houses or roads and health care.

(2) Law-finding sciences versus fact-finding sciences.

chemistry and physics attempt to discover universal laws which are applicable
everywhere at all times
geography, history (if it is a science), and perhaps economics are concerned
with local events. It is often said that the subject matter of the latter consists of
particular facts, not general laws.

As a result, there are some who wish to limit the term `science' to the law-finding sciences.
Upon the basis of the criteria of science (such as those which will be presented later, or
others), we believe that we may say that both the law-finding disciplines and the fact-finding
disciplines are capable of being sciences if those (or other) criteria are met. Furthermore, one
2
might argue that there are no purely fact-finding sciences. So, to speak of law-finding versus
fact-finding may, in many cases, indicate at artificial disjunction.

(3) Natural sciences versus social sciences.

Sometimes the distinction is based on subject matter.
Hence: natural phenomena constitute the field of science
cultural phenomena constitute the field of scholarship and require
understanding, verstehen, and empathy.

But there are points at which the classification does not hold up.
First, there are some predominantly fact-finding natural sciences,
such as geography, geology, and palaeontology.
And there are some law-finding social sciences,
such as sociology and linguistics.
Second, the distinction according to subject matter is not a clear-cut one.

Hence we shall take a "liberal" view of science and allow the use of the term
science to apply to both the natural and the social sciences--with the recognition
that there are some differences.


II. THE AIMS OF SCIENCE
Using the above distinction between pure (empirical) and applied science, we
may cite the following as some of the aims of science.

(1) The aims of applied science include: control, planning, technological progress;
the utilization of the forces of nature for practical purposes.
Examples are: flood control, the construction of sturdier bridges, and the
improvement of agriculture.

(2) The aims of the pure, factual sciences may be considered from two standpoints.
(a) Psychologically considered, the aims of the pure, empirical sciences are:
the pursuit of knowledge;
the attainment of truth (or the closest possible realization of truth);
the satisfaction of using our intellectual powers to explain and predict
accurately.
Scientists, of course, derive enjoyment from rewards, prestige, and competing with
others. But they often achieve a genuine inner gratification which goes with the search for
truth. In some ways this is similar in quality to artistic satisfaction. It is seen, for example,
in the enjoyment one derives from the solution of a difficult problem.

(b) Logically considered, the aims are often held to be:
description, explanation, and prediction.
(b1) Description includes giving an account of what we observe in certain
contexts, the formulation of propositions which apply to (or correspond
to) facts in the world.
3
(b2) Explanation consists of accounting for the facts and regularities we
observe. It involves asking and answering `Why?' or `How come?' This
may be done by subsuming facts under laws and theories.
(b3) Prediction is closely related to explanation. It consists in deriving
propositions which refer to events which have not yet happened, the
deducing of propositions from laws and theories and then seeing if they
are true, and hence provide a testing of those laws and theories.
(b4) We might also mention post- or retro-diction, the reconstruction of past
events. This process is also inferential in character.

III. THE CRITERIA OF SCIENCE
ESSENTIAL CRITERIA of science which may be used for at least two purposes:
First, to distinguish science from commonsense knowledge (without claiming that
the two are radically disjunctive--in some cases that may differ only in degree, not in
kind);
Second, to distinguish that which is scientific, on the one hand, from that which is
either nonscientific or unscientific, on the other
- for example, between theories which are genuinely scientific and those which
are not. An enterprise, discipline, or theory is scientific if it is characterized by or
meets those criteria.

These criteria have been stated by Professor in various lectures Herbert Feigl
and in writing (see selection 28 in Klemke, et. al., 1988). (His is only one view.)

The five criteria are:
(1) Intersubjective testability. This refers to the possibility of being, in principle,
capable of corroboration or "check-up" by anyone. Hence: intersubjective.
(Hence, private intuitions and so forth must be excluded.)
(2) Reliability. This refers to that which, when put to a test, turns out to be true, or at
least to be that which we can most reasonably believe to be true. Testing is not
enough. We want theories which, when tested, are found to be true.
(3) Definiteness and precision. This refers to the removal of vagueness and
ambiguity. We seek, for example, concepts which are definite and delimited. We
are often helped here by measurement techniques and so forth.
(4) Coherence or systematic character. This refers to the organizational aspect of a
theory. A set of disconnected statements is not as fruitful as one which has
systematic character. It also refers to the removal of, or being free from,
contradictoriness.
(5) Comprehensiveness or scope. This refers to our effort to attain a continual
increase in the completeness of our knowledge and also to our seeking theories
which have maximum explanatory power-for example, to account for things
which other theories do not account for.




4
Let us consider these criteria in greater detail.

(1) Intersubjective testability.
(a) Testability. We have noted that in science we encounter various kinds of statements:
descriptions, laws, theoretical explanations, and so on. These are put forth as knowledge-claims. We
must (if possible) be able to tell whether evidence speaks for or against such knowledge claims. If the
propositions which express those claims are not capable of tests, we cannot call those propositions true
or false or even know how to go about establishing their truth or falsity. It should be noted that the
criterion is one of testability, not tested. For example, at a given point in time, `There are mountains on
the far side of the moon' was testable though not tested.
(b) Intersubjective. 'Intersubjectivity' is often employed as a synonym for `objectivity.' And
the latter term has various meanings. Some of these are:
(i) A view or belief is said to be objective if it is not based on illusions,
hallucinations, deceptions, and so on.
(ii) Something is referred to as objective if it is not merely a state of mind but
is really "out there" in the external world.
(iii) We often use `objective' to indicate the absence of bias and the presence
of disinterestedness and dispassionateness.
(iv) `Objectivity' also refers to the possibility of verification by others, and
hence excludes beliefs, which stem from private, unique, unrepeatable
experiences. Science strives for objectivity in all of these senses. Hence Feigl takes
'intersubjective' to include all of them.

(c) Intersubjective testability. It is often held that (according to the view we are
considering) in order for a proposition or theory to be judged scientific it must meet this first
requirement. Indeed, many of the other criteria presuppose intersubjective testability. We cannot
even begin to talk of reliability or precision unless this first criterion has been met.

(2) Reliability. Science is also interested in those which are true or at least have the greatest
verisimilitude or likelihood of being true. Hence the need arises for the criterion of reliability.
Whereas the first criterion stressed the possibility of finding assertions which are true or false, the
second stresses the end result of that process. We judge a claim or body of knowledge to be
reliable if it contains not merely propositions which are capable of being true or false but rather
those which are true or which have the greatest verisimilitude. We find such propositions to be
true (or false) by means of confirmation. Complete verification, and hence complete certainty,
cannot be achieved in the factual sciences.
It should be noted that, first, the reliability of scientific assertions make them useful for
prediction; second, although the assertions of many enterprises are testable (for example, those
of astrology as much as those of astronomy), only some of them are reliable. And we reject some
of them precisely because they are unreliable. The evidence is against them; we do not attain
truth by means of them.

(3) Definiteness and precision. The terms 'definiteness' and `precision' may be used in
at least two related senses.
First, they refer to the delimitation of our concepts and to the removal of
ambiguity or vagueness.
Second, they refer to a more rigid or exact formulation of laws. For example, `It
is more probable than not that X causes disease Y' is less desirable than `The
probability that X causes Y is 98%.'
5

(4) Coherence or systematic character. In the sciences, we seek a well-connected
account of the facts. We achieve this via what has been called the hypothetico-
deductive procedure of science. This procedure includes:
(a) our beginning with a problem (which pertains to some realm of phenomena);
(b) the formulation of hypotheses, laws, and theories by which to account for
those phenomena or by which to resolve the problem;
(c) the deriving (from (b)) of statements which refer to observable facts;
(d) the testing of those deduced assertions to see if they hold up.

Thus we seek an integrated, unified network, not merely a congeries of true statements. But, of
course, we also seek theories which are consistent, which are free from self-contradictions. The
reason for insisting upon such coherence is obvious; hence there is no need for elaboration.

(5) Comprehensiveness or scope. The terms 'comprehensiveness' and `scope' are
also used in two senses:
First, a theory is said to be comprehensive if it possesses maximum
explanatory power. Thus Newton's theory of gravitation was ranked high partly
because it accounted not only for the laws of falling bodies but also for the revolution of
the heavenly bodies and for the laws of tides.
Second, by 'comprehensive' we often refer to the completeness of our
knowledge. This of course does not mean finality. We do not think of the hypotheses of
the empirical sciences as being certain for all time. Rather we must be ready to modify
them or even, on occasion, to abandon them.

To summarize: According to the view we have presented, we judge a law, hypothesis,
theory, or enterprise to be scientific if it meets all five of the above criteria. If it fails to meet all
five, it is judged to be unscientific or at least nonscientific. To return to our earlier example, it
seems clear that Newton's theory thus passes the test. Astrological theory or Greek mythology
does not.
It should be noted that, in presenting Professor Feigl's criteria for the reader's consideration,
we do not claim that they are correct or free from defects. Indeed, as we shall see in the
readings which follow, many writers have rejected some (or all) or those criteria. The reader
should once again attempt to seek an acceptable criterion or set of criteria, if such can be found.

6
IV. WHAT IS SCIENCE?

A common characterization of science (or sometimes of scientific method) runs as
follows. Science is knowledge obtained by: (1) making observations as accurate and
definite as possible; (2) recording these intelligibly; (3) classifying them according to the
subject matter being studied; (4) extracting from them, by induction, general statements
(laws) which assert regularities; (5) deducing other statements from these; (6) verifying
those statements by further observation; and (7) propounding theories which connect
and so account for the largest possible number of laws. It is further maintained that this
process runs from (1) through (7) in that order.
The conception of science has been challenged in recent years. Its most severe critic is
Sir Karl Popper. (See the selection in Part 1 of this volume.) We, shall not repeat
Popper's criticisms. Instead we offer a characterization of science which some believe to
be more adequate than the one mentioned above and which they deem to be free from
the defects it possesses.

The following is at least a minimal characterization of a science

Science is a body of knowledge which consists of the following, coherently
organized in a systematic way:
(a) Statements which record and classify observations which are relevant for the
solution of a problem in as accurate and definite a way as possible.
(b) General statements-laws or hypotheses-which assert regularities among certain
classes of observed or observable phenomena.
(c) Theoretical statements which connect and account for the largest possible
number of laws.
(d) Other general or specific statements which are deducible from the initial
descriptions and from laws and theories and which are confirmed by further
observation and testing.

At least two things should be noted about this characterization.
First, it indicates the role of the formal sciences in the empirical sciences.
Mathematics is important for (a); logic is important for (d).
Second, nothing is said in this characterization about the method of obtaining
knowledge or of obtaining laws. It may be induction, but it may also be a guess,
intuition, hunch, or whatever.