Philosophy of Science and Economics:
Positivism, Popper, and After
A Story in Four Parts
1. The Philosophy of Science Background
a. Positivism (Logical Positivism and
Logical Empiricism)
b. The Popperian Tradition
2. Two Mainstreams: Economics and
Philosophy of Natural Science in the Mid20th Century
3. Economic Methodology Then (1980s &
1990s) and Now (recent developments)
Philosophy of Natural Science in the 20th
"If it is true that there are but two kinds of people in the world the logical positivists and the god-damned English professors then I suppose I am a logical positivist." [Glymour, 1980, ix]
"Once, in those dear dead days, almost, but not quite beyond
recall, there was a view of science that commanded
widespread popular and academic assent. This view deserves
a name. I shall call it "Legend." ... So much for the dear dead
days. Since the 1950s the mists have begun to fall. Legend's
lustre has dimmed. While it may continue to figure in textbooks
and journalistic expositions, numerous intelligent critics now
view Legend as smug, uninformed, unhistorical, and
analytically shallow." [Kitcher, 1993, pp. 3-5]
Logical Positivism and Logical
Three Periods of "Positivism"
• 19th Century Positivism
• Logical Positivism: Primarily 1930s
• Logical Empiricism: mainstream
philosophy of natural science 1940-1980
(and continuing on in a slightly modified
way today – at least as the obligatory
point of reference)
1a. Positivism (Logical Positivism and
Logical Empiricism)
Logical Positivism: Associated with the
Vienna circle 1920-35. Key figures
include Schlick, Carnap, Reichenbach,
Neurath, Ayer (in UK), Wittgenstein
(early), … Self-consciously scientistic
(with narrow conception of "science").
Goal to replace all philosophical
"knowledge" with knowledge in the image
of science.
Two types of meaningful discourse (only two
1. Analytical Knowledge: mathematics and logic.
Nonempirical, a priori.
2. Scientific Knowledge: Science (particularly
natural science). Empirical and a posteriori.
Meaningless of Metaphysics (and everything else
not in the previous two categories).
Verifiability criterion of meaningfulness: for
a sentence to be meaningful it must be
"in principle verifiable," that is, it must
be possible to specify, at least in
principle, the conditions under which the
sentence would be true; "observational
evidence can be described which, if
actually obtained, would conclusively
establish the truth of the sentence"
(Hempel, 1965, p. 103).
Basic Logical Positivist View of Scientific
Theories (Four parts)
1. Logical vocabulary and formal rules
of inference (logic, mathematics, set
theory, …)
2. Theoretical vocabulary (VT)
consisting of theoretical terms
(force, mass, gene, friction, electron,
price level, real output, utility
function, profit, …)
3. Observation vocabulary (V0)
consisting of terms that are directly
observable (the empirical basis of
4. Correspondence Rules (C)
relating/translating each of the
theoretical term in VT to the
observational language V0. Each
theoretical term will have an explicit
definition – will correspond to – in
the observational vocabulary.
Logical Empiricism = Received View = Legend (RevisedReformed, 2nd generation logical positivism): Richard
Braithwaite, Carl Hempel, Ernest Nagel, and others. Most
work done in the U. S. by philosophers who had studied
in/with the Vienna Circle. Academic professionalization of
"fields" within philosophy.
Hempel, Carl G. (1965), Aspects of Scientific Explanation. N. Y.: The Free Press.
Hempel, Carl G. (1966), Philosophy of Natural Science. Englewood Cliffs, NJ:
Nagel, Ernest (1961), The Structure of Science: Problems in the Logic of Scientific
Explanation. N. Y.: Harcourt, Brace & World.
Salmon, Wesley C. (1966), The Foundations of Scientific Inference. Pittsburgh, PA:
University of Pittsburgh Press.
Many many small changes from logical
positivism, but the three most important
• Gradual Erosion of the Theory/Observation
• Realist vs. Instrumentalist View of Theories
• D-N Model of Scientific Explanation
"Furthermore, there remains no satisfactory general way of dividing all
conceivable systems of theoretical terms into two classes: those that
are scientifically significant and those that are not; those that have
experiential import and those that lack it. Rather, experiential, or
operational, significance appears as capable of gradations. ...
experiential significance presents itself as capable of degrees, and any
attempt to set up a dichotomy allowing only experientially meaningful
and experientially meaningless concept systems appears as too crude
to be adequate for a logical analysis of scientific concepts and theories."
(Hempel, 1965, pp. 131-32)
Debate over the Cognitive Status of Scientific Theories: The
instrumentalist view of theories versus the realist view of theories
In its simplest form "instrumentalism" says that scientific theories are
merely instruments for making empirical predictions; on the other
hand, the simplest version of scientific "realism" says that scientific
theories contain statements that can actually be true or false.
Scientific Explanation (Redescripton in Early Logical
Positivism vs. Explanation in Logical Empiricism)
Deductive-Nomological (or D-N) model of scientific explanation.
Schematically then, a D-N explanation will take the following
general form.
C 1, C 2, . . . C n
L1, L2, . . . Lm
where each Ci represents a sentence that describes an initial
condition, and each Li represents a general law.
1b. Karl Popper and the Popperian
Philosophical Tradition
Karl R. Popper (1902-1994)
Logik der Forschung (1934)[English
translation as The Logic of
Scientific Discovery (1968)]
The Poverty of Historicism (3rd ed.
Conjectures and Reputations (2nd ed.
The Open Society and Its Enemies
(2nd ed. 1966)
Objective Knowledge (1972)
Unended Quest: An Intellectual
Autobiography (1976)
Realism and the Aim of Science (1983)
The Myth of the Framework (1994)
Popperian Falsificationism: Popper's story about Marx and Freud.
Confirmation is easy. Good science involves risk. A scientific theory will
stick its neck out!
Falsification (not confirmation): Popper's "solution" to the problem of
Induction: Falsification not Confirmation. There is no Induction in Science.
Testability is falsifiability.
Modus Ponens
Therefore B
Modus Tollens
Therefore ~A
Demarcation Criterion between Science and non-Science: For a theory to be
"scientific" it must be potentially falsifiable by at least one empirical
basis/basic statement.
The Growth of Knowledge
Corroboration/Testability: What if nonfalsified? Choose the
more severely tested (most corroborated) or the more
testable (most potentially falsifiable). The existing body of
scientific knowledge is all falsifiable, but not falsified
scientific theories.
Bold Conjecture and Severe Tests
Conjectures and Refutations
Novel Facts/ Non-Ad-Hocness
The problem of Ad-hoc theory adjustments. Avoid ad-hoc
theory adjustments through novel facts/novel
Popper on Social Science/Economics (Situational AnalysisSA and the Rationality Principle-RP)
"My views on the methodology of the social sciences are the result of my
admiration for economic theory: I began to develop them, some
twenty-five years ago, by trying to generalize the method of
theoretical economics." (Popper, 1994, p. 154
SA explanation of why agent A did act X can be given in the
following schematic form.
Description of the
Agent A was in
situation S.
Analysis of the
In situation S the
appropriate (rational)
thing to do is X.
The Rationality
Principle (RP):
Agents always act
(rationally) given their
Therefore: A did X.
2. Two Mainstreams: Economics and
Philosophy of Natural Science in the
Mid-20th Century
The Question here (Hands 2007) is: Why did
mainstream economics look so much like
mainstream philosophy of natural science
during the period (roughly) 1945-1965?
Two Parts:
1. Show that in fact the did "look so much alike"
2. Suggest some of the reasons why that may
have been the case (just a start on a much
more detailed historical investigation)
The Similarities are Uncanny
Both stabilized during this time: they moved from interwar
diversity to a single dominant paradigm in the postwar era
(neoclassical synthesis in economics and logical empiricism
in philosophy of natural science.
The stabilization in both fields involved a formalist
Rationality: Both fundamentally concerned with rationality.
It is economic rationality in the case of economics and
scientific rationality in the case of the philosophy of science.
Economic rationality is a version of practical rationality
(what it is rational to do) while in philosophy of science it is
epistemic rationality (what it is rational to believe), but both
focus on rationality.
Both have a positive/descriptive aspect: economics is
concerned with the prediction and explanation of economic
behavior while philosophy of science is concerned with
describing the practice of the best scientists.
The Similarities are Uncanny (continued)
Both also have normative aspects. In economics there are two: at
the individual level rational choice theory tells agents what they
ought to do in order to be rational, and at the social level the
theory supports an ethically normative welfare economics.
Philosophy of natural science also involves normativity at both the
individual and social levels: at the individual level it tells
scientists what they ought to do to produce scientific knowledge
and a the level of the scientific community it guides science policy
and institutional design.
After 1965 or so both were challenged wrt descriptive inadequacy.
In economics Keynesian theory was empirically challenged during
the 1970s and microeconomics was challenged later by
experimental & behavioral economics, while the challenge to
philosophy of science came from Kuhn and others.
Both reconciled empiricism/prediction and realism/explanation
in ways that relieved early tensions between these two aspects of
the programs.
Okay so why?
The Viennese Connection: origins of both logical
empiricism and mathematical economics
(Walrasian general equilibrium, game theory, and
2. The Harvard Connection (Science of Society
Discussion Group, Samuelson's writing Vanevar
Bush's Science, the Endless Frontier.
3. The Operations Research Connection (Mirowski's
story). Both sought to reconcile military inspired
"command and control" with democracy in the
context of the cold war.
These are just the tip of the historical iceberg.
3. Economic Methodology Then and Now
Changes in Economic Methodology During
the Last 35 (or so) Years
The development of a “field”
The changing relationship between
orthodox and heterodox economics
Recent changes in macroeconomics versus
recent changes in microeconomics
A two-part story:
The period (roughly) 1975-2000
Recent developments
Two Exemplary Economic Methodologists of the
period 1975-2000: Mark Blaug (1927-2011) and
Terence Hutchison (1912-2007)
“I argue in favor of falsificationism, defined as a
methodological standpoint that regards theories
and hypotheses as scientific if and only if their
predictions are at least in principle falsifiable, that
is, if they forbid certain acts/states/events from
occurring … In addition, I claim that modern
economists do in fact subscribe to the methodology
of falsificationism: … I also argue, however, that
economists fail consistently to practice what they
preach: their working philosophy of science is aptly
characterized as ‘innocuous falsificationism’.”
(Blaug, 1992, p. xiii).
Common Features of the (Popperian)
Methodology of Blaug and Hutchison
Very Critical of Heterodox Economics because it
Failed to Make Bold Conjectures and Expose them
to Severe Empirical Tests
Also Very Critical of Neoclassical (particularly
Walrasian) Economics because it Failed to Make
Bold Conjectures and Expose them to Severe
Empirical Tests
Bottom Line: Economic Methodologies so “tough” that
nothing – Heterodox or Neoclassical – Passed the
Test. But also No Practical Guidance for how to
Improve Practice. No Constructive, only
Destructive, Criticism.
The Recent Literature: The Debate Over
Whether there is New More Pluralistic
Mainstream Debate
Has Experimental Economics, Behavioral Economics,
Neuroeconomics, Complexity Theory, and Other
Recently Expanding Fields within (micro) Economics
Changed the Mainstream?
“Economics is Moving Away from Strict Adherence to
the Holy Trinity – Rationality, Selfishness, and
Equilibrium – to a More Eclectic Position of
Purposeful Behavior, Enlightened Self-Interest, and
Sustainability” (Colander, Holt, and Rosser, 2008, p.
Maybe, or maybe not, but there is a tendency in that
direction, and the research is neither neoclassical
nor heterodox.
The Fields Identified as the New More
Pluralistic Mainstream Certainly Are Where
Most of the Recent Methodological Research is
Being Done
Significant Texts in Recent Methodology:
Bardsley et al. (2010), Guala (2005), Reiss
(2007), Ross (2005), Santos (2010), and
Others …
Example of The Elgar Companion to Recent
Economic Methodology
The Practitioner’s View: Gul and Pesendorfer
The Lessons From This Recent Literature
The Majority of Recent Methodological Research is in
these New Pluralistic (micro) Economic Fields
The Theory Being Methodologically Investigated is
Neither Strictly Neoclassical Economics Nor
Traditional Heterodox Economics
The Literature is Less Universalistic, More Local and
Naturalistic, and More Sensitive to the Practical
Concerns and Specific Features of the Relevant
It is Particularistic, but it is Often Critical, and is
Based on Philosophical Justification and

Philosophy of Science and Economics: Positivism, Popper