Wetenschapsfilosofie lectures
Lecture 1
Philosophy of social science: thinking hard about social science
Philosophy of social science is about reflecting on social science itself, in order to
understand it better
Logical positivism:
- Vianna Circle: group of scientists (in early 20 th century Vienna) reflecting on philosophical
questions about science
- Aim: development of a strictly scientific worldview
- Against speculative philosophy, religious ideas, traditional worldviews
Ideals:
- Strict empiricism: only empirical observation can give us knowledge
o No place for speculative and theoretical claims that are not based on observation
- Use of formal logic and mathematics to create an ideal and precise language for science
o To guard against unwarranted terminology and against leaps to conclusions and
unsupported theories
Core ideas: verifiability
- An ideal and precise language of science
o Gate-keeping: only statements that are firmly based on empirical observation belong
in the language of science
o The verifiability criterion of meaning: the meaning of a statement is its method of
verification
Core ideas: demarcation
- Verifiability as demarcation criterion
o Only statements that satisfy the verifiability criterion are scientific, other statements
are non-scientific
o (logic, mathematics, and statistics are not verifiable, but they are merely linguistic
conventions to help formulate scientific statements in a precise manner)
Core ideas: induction
- Inductive method: from observations to general theories and empirical regularities/laws
o Observations give rise to hypotheses and theories
o And they serve to support/confirm them
- Let the data (observations) ‘speak for themselves’
,Example of logical positivism applied to social science: behaviorism
Behaviorism: exclusive focus on observable behavior in response to external stimuli
- Nothing about internal cognitive processes
- Because those are unobservable/unverifiable
Karl Popper’s core ideas:
- Fallibility and tentativeness of human knowledge
- Dogmatic vs. critical thinking
Problem of induction:
- Reasoning from individual observations to general conclusions is logically invalid
- So induction can never completely support general scientific laws and theories
- Popper: no use for induction in science
Falsifiability as demarcation (prove that something is true or not)
- Scientific knowledge is falsifiable knowledge
o Unicorns exist and don’t exist
- Scientific statements ought to be able to ‘clash’ with the world
o Unicorns either exist or they don’t
- It must be possible to prove them false through experiments and observations
o Unicorns don’t exist
Examples of unfalsifiable theories
- Freud: every little boy has an Oedipus complex, or is in denial of it
- Marx: changes in the means of production lead to changes in labor conditions, which lead to
changes in political power, which in turn lead to changes in ideology
,Example of a falsified theory: secularization thesis (popular in 19 th – 20th century sociology)
- Through enlightenment modernization, rationalization, combined with the ascent of science
and technology, religious authority diminishes in all aspect of social life and governance
Scientific method for Popper:
- Science is about formulating theories (conjectures) in such a way that they can be falsified by
empirical observation
- Theories must then be tested as rigorously as possible (attempted refutations)
- We accept those theories that have survived testing (so far)
Comparison: Popper vs. logical positivism
Popper Logical positivism
Fallibility and risk-taking Striving for certainty
Theoretical conjectures as starting points Observations as starting points
Get rid of bad ideas as you go Don’t led any bad ideas in
Lecture 2
Science ought to be value-free
Two methods to count votes: door-to-door & mail-in questionnaires
Both methods have their pros and cons. Values and reasons for choosing a method:
- Door to door: depends on people wanting to talk to government rep; more accurate for
people who are hard to locate; but expensive and time-consuming
- Questionnaire: depends on effort to mail in; cheaper
Kinds of values:
- Epistemic values: values that have to do with truth and knowledge
- Non-epistemic values: values that don’t have to do with truth and knowledge (political ideas,
social values, worldviews)
The influence of epistemic values in science is unproblematic
Different roles for values:
- Contextual role: when values are involved in the context in which research takes place, but
not necessary for conducting research
- Constitutive role: when values constitute an activity, are necessary to the activity of research,
shape it from the inside (values involved in choice of methods, data gathering, analyses,
drawing conclusions
Contextual role for non-epistemic values: what research to fund, what problems to investigate, where
and how to communicate about finding, with whom are results shared etc.
Constitutive role for epistemic values: what method gives the most accurate results, which analysis
technique is best, is a sample representative, what to do with outliers, etc.
, Moderate thesis of value freedom: science is objective when only epistemic values are constitutive of
scientific practice; moral and political values must always remain contextual.
- But its more complicated..
- Even the moderate thesis of value freedom faces problems
- Because non-epistemic values often play a constitutive role in scientific practice, and
inevitably so
Values in science 1: weighing different kinds of errors
- Which type of error matters more: overcounting or undercounting the population
- More generally: which type of error is worse?
o Type 1 error: false positive, concluding there is some effect when there is none
o Type 2 error: false negative, concluding there is no effect when there is one
- Deciding what type of error matters more is never a scientific decision, but always a value-
laden one
- Non-epistemic values play a constitutive role here
Values in sciences 2: definition of concepts
- Who is a ‘citizen’?
- Defining a legal status like citizen is not purely scientific issue, but also depends on value
judgements
- The same goes for many concepts in social science: labor, sex, gender, inequality, etc.
- Since defining and operationalizing concepts is part of scientific practice, this is another
example where non-epistemic values must play a constitutive role in science
Values in science 3: methods and morality
- Moral reasons for or against the use of research methods
o Milgram experiment
o Animal testing
o Deception, manipulation
Values in science 4: emancipatory and critical research
- Social science can itself be oppressive and exclusionary by working from a privileged
perspective or political ideology
- Recognizing this and changing it requires commitment to political values
- Concepts like injustice, oppression, equal, prejudice, etc. are both descriptive and normative
(value laden)
- Thick moral/evaluative concepts
- Such concepts are inevitable for certain social science research
- Hence, a constitutive role for non-epistemic values
Value-ladenness and objectivity?
- Doesn’t constitutive involvement of non-epistemic values threaten the objectivity of social
science?
Lecture 1
Philosophy of social science: thinking hard about social science
Philosophy of social science is about reflecting on social science itself, in order to
understand it better
Logical positivism:
- Vianna Circle: group of scientists (in early 20 th century Vienna) reflecting on philosophical
questions about science
- Aim: development of a strictly scientific worldview
- Against speculative philosophy, religious ideas, traditional worldviews
Ideals:
- Strict empiricism: only empirical observation can give us knowledge
o No place for speculative and theoretical claims that are not based on observation
- Use of formal logic and mathematics to create an ideal and precise language for science
o To guard against unwarranted terminology and against leaps to conclusions and
unsupported theories
Core ideas: verifiability
- An ideal and precise language of science
o Gate-keeping: only statements that are firmly based on empirical observation belong
in the language of science
o The verifiability criterion of meaning: the meaning of a statement is its method of
verification
Core ideas: demarcation
- Verifiability as demarcation criterion
o Only statements that satisfy the verifiability criterion are scientific, other statements
are non-scientific
o (logic, mathematics, and statistics are not verifiable, but they are merely linguistic
conventions to help formulate scientific statements in a precise manner)
Core ideas: induction
- Inductive method: from observations to general theories and empirical regularities/laws
o Observations give rise to hypotheses and theories
o And they serve to support/confirm them
- Let the data (observations) ‘speak for themselves’
,Example of logical positivism applied to social science: behaviorism
Behaviorism: exclusive focus on observable behavior in response to external stimuli
- Nothing about internal cognitive processes
- Because those are unobservable/unverifiable
Karl Popper’s core ideas:
- Fallibility and tentativeness of human knowledge
- Dogmatic vs. critical thinking
Problem of induction:
- Reasoning from individual observations to general conclusions is logically invalid
- So induction can never completely support general scientific laws and theories
- Popper: no use for induction in science
Falsifiability as demarcation (prove that something is true or not)
- Scientific knowledge is falsifiable knowledge
o Unicorns exist and don’t exist
- Scientific statements ought to be able to ‘clash’ with the world
o Unicorns either exist or they don’t
- It must be possible to prove them false through experiments and observations
o Unicorns don’t exist
Examples of unfalsifiable theories
- Freud: every little boy has an Oedipus complex, or is in denial of it
- Marx: changes in the means of production lead to changes in labor conditions, which lead to
changes in political power, which in turn lead to changes in ideology
,Example of a falsified theory: secularization thesis (popular in 19 th – 20th century sociology)
- Through enlightenment modernization, rationalization, combined with the ascent of science
and technology, religious authority diminishes in all aspect of social life and governance
Scientific method for Popper:
- Science is about formulating theories (conjectures) in such a way that they can be falsified by
empirical observation
- Theories must then be tested as rigorously as possible (attempted refutations)
- We accept those theories that have survived testing (so far)
Comparison: Popper vs. logical positivism
Popper Logical positivism
Fallibility and risk-taking Striving for certainty
Theoretical conjectures as starting points Observations as starting points
Get rid of bad ideas as you go Don’t led any bad ideas in
Lecture 2
Science ought to be value-free
Two methods to count votes: door-to-door & mail-in questionnaires
Both methods have their pros and cons. Values and reasons for choosing a method:
- Door to door: depends on people wanting to talk to government rep; more accurate for
people who are hard to locate; but expensive and time-consuming
- Questionnaire: depends on effort to mail in; cheaper
Kinds of values:
- Epistemic values: values that have to do with truth and knowledge
- Non-epistemic values: values that don’t have to do with truth and knowledge (political ideas,
social values, worldviews)
The influence of epistemic values in science is unproblematic
Different roles for values:
- Contextual role: when values are involved in the context in which research takes place, but
not necessary for conducting research
- Constitutive role: when values constitute an activity, are necessary to the activity of research,
shape it from the inside (values involved in choice of methods, data gathering, analyses,
drawing conclusions
Contextual role for non-epistemic values: what research to fund, what problems to investigate, where
and how to communicate about finding, with whom are results shared etc.
Constitutive role for epistemic values: what method gives the most accurate results, which analysis
technique is best, is a sample representative, what to do with outliers, etc.
, Moderate thesis of value freedom: science is objective when only epistemic values are constitutive of
scientific practice; moral and political values must always remain contextual.
- But its more complicated..
- Even the moderate thesis of value freedom faces problems
- Because non-epistemic values often play a constitutive role in scientific practice, and
inevitably so
Values in science 1: weighing different kinds of errors
- Which type of error matters more: overcounting or undercounting the population
- More generally: which type of error is worse?
o Type 1 error: false positive, concluding there is some effect when there is none
o Type 2 error: false negative, concluding there is no effect when there is one
- Deciding what type of error matters more is never a scientific decision, but always a value-
laden one
- Non-epistemic values play a constitutive role here
Values in sciences 2: definition of concepts
- Who is a ‘citizen’?
- Defining a legal status like citizen is not purely scientific issue, but also depends on value
judgements
- The same goes for many concepts in social science: labor, sex, gender, inequality, etc.
- Since defining and operationalizing concepts is part of scientific practice, this is another
example where non-epistemic values must play a constitutive role in science
Values in science 3: methods and morality
- Moral reasons for or against the use of research methods
o Milgram experiment
o Animal testing
o Deception, manipulation
Values in science 4: emancipatory and critical research
- Social science can itself be oppressive and exclusionary by working from a privileged
perspective or political ideology
- Recognizing this and changing it requires commitment to political values
- Concepts like injustice, oppression, equal, prejudice, etc. are both descriptive and normative
(value laden)
- Thick moral/evaluative concepts
- Such concepts are inevitable for certain social science research
- Hence, a constitutive role for non-epistemic values
Value-ladenness and objectivity?
- Doesn’t constitutive involvement of non-epistemic values threaten the objectivity of social
science?
