Reading 2: ch.2&3 (philosophy of social science) Flashcards
1
Q
positivism
A
- movement to establish a sound basis for social scientific inquiry = 1950s and 1960s (behavioural revolution)
- maintains that researchers can arrive at factual, reliable and objective answers to questions about the social world by employing the methods used in the natural sciences
- claims (just like scientific realism) that the social world is no different than the natural world
2
Q
(positive political theory)
A
assumes that rational self-interest motivates behavior
NOT attitudes
3
Q
behavioral revolution changed the field:
A
- new discussions and debate about desirability and possibility of using scientific methods to attain reliable, empirical, causal knowledge
- broadening the domain of political research (by drawing on theories from other disciplines)
- emphasis on research based on empirical observation (falsifiable)
- importance of replication
4
Q
classical positivism
- tenets
- about causality
A
4 basic tenets:
- naturalism (no difference social and natural sciences/world)
- empiricism (what we know of the world is limited to that what we can see)
- goal of social science is to explain and predict by means of laws, which can be established based on induction
- it is possible to distinguish between facts and values + to obtain value-free knowledge
causality conceptualization of Hume (causality constituted by empirical regularities among observable variables, this leads to a psychological expectation of causality, but we cannot see it therefore not know it) -> seeks empirical regularities rather than to discover causal mechanisms
5
Q
logical positivism
A
- early 20th century as movement within philosophy
contributes to positivist thought:
- wants to combine induction (empiricism) and deduction (logic): argues that logical reasoning and mathematics should also be treated as sources of knowledge in addition to empiricism
- establishes verification as criterion for establishing truth claims + science (rather than metaphysics)
*some argue that its greatest contribution to positivism is that it inspired critique from Popper
6
Q
deduction vs induction
+ retroduction
A
induction = particular observations/cases -> generalizations
- observation -> pattern - tentative hypothesis -> theory
deduction = broad generalizations/theories -> specific observations and meanings/implications
- theory -> hypothesis -> observation -> confirmation
in practice research is often retroduction: interaction of induction and deduction in an evolving, dynamic process of discovery and hypothesis formation
7
Q
Popper
A
against (logical positivist) principle of verification: we cannot deductively establish general statements of scientific knowledge
- induction is bad: no matter how many observations confirm a theory, only one observation is necessary to falsify it
we shouldn’t try to verify a hypothesis, we should try to falsify it
falsification = demarcation science and pseudo-science/metaphysics
science/theory doesn’t start with observation, observation is used to test and falsify theory
8
Q
criticism to Popper
A
- he distinguishes between fact and theory (he claims that our observations/facts can be established independently of the theory that they mean to test)
- Popper’s notion of falsifiability is at odds with how scientists work in practice (they don’t really seek to falsify their theories)
9
Q
classical positivism vs logical positivism vs Popper’s critique
A
classical positivism: science through induction
logical positivism: induction + deduction can be used to discover laws
Popper: only deduction to establish laws of social life as a basis for explanation
10
Q
deductive-nomological model
A
Hempel
something is explained when it is shown to be a member of a more general class of things, when it is deduced from a law or a set of laws
when is something a law rather than it appears to be a law but is accidental?
law expresses a necessary connection between properties, accidental generalization doesn’t + laws can be tested based on their predictions
we confirm that a generalization is a law by treating it as a hypothesis (hypothetico-deductive model)
11
Q
hypothetico-deductive model
A
Hempel
according to this model we confirm that a generalization is a law by treating it as a hypothesis, testing the hypothesis by deducing it from predictions of further phenomena that should be observable as a consequence of the hypothesis
*we look at multiple hypotheses to see which has the most explanatory value (look which hypothesis makes the most accurate predictions)
12
Q
scientific realism
A
- naturalist ontology (social world and natural world are the same) = same as positivism
- what is objectively real is not just observable elements, also unobservable elements (we can see its effects) = breaks with positivism
- explanation can be based on observable regularity AND unobservable causal mechanisms that link cause and effect
- ## scientific goal = describe and explain observable and nonobservable aspects of the world
conception of causality = observable regularities + unobservable causal mechanisms that generate the regularities
13
Q
Charles Tilly - 3 causal mechanisms in the social world
A
*mentioned with scientific realism
- environmental (external influences on conditions effecting social life)
- cognitive (operates through alterations of individual and collective perception)
- relational (alter connections among people, groups and interpersonal networks)
14
Q
debate about the scientific status of unobservable entities
- 3 main questions
A
- What is the ontological status of macro-social mechanisms used to explain social outcomes? = What are the basic entities that make up the social world?
- (methodological) individualism (social phenomena are made up of combined results of individual action)
- (methodological) holism (social facts have social causes that are irreducible to facts about individuals: the whole is not directly explicable in term of its parts) - How do we explain macro-social phenomena?
- methodological individualism vs methodological holism (problem = reification -> not enough focus on agency) vs ‘micro-foundations (combination individualism and holism: intentional states that motivate individual action) - How do macro-social ‘social mechanisms’ produce social outcomes? Providing explanations of macro-social phenomena with micro-foundations
- James Coleman: diagram (Coleman’s Bathtub/boat): causal relations flow downwards from macro phenomena (e.g. institutions) shape the conditions of individual actions + conditions give rise to individual actions + individual actions aggregate up to macro outcomes
- he argues for a combination of micro-macro linkages (see p. 45)
15
Q
reification
A
tendency to treat macro-social structural entities as if they had a concrete, material existence; to treat them as analytically indepe3ndent of their constituent elements; inert, unchanging and unmediated by human agency
*problem with methodological holism
e.g. globalization as result of capitalism and market thinking
16
Q
critical realism
A
- reality consists of observable and unobservable elements (knowable through their effects) = same as scientific realism
- is critical: sees opportunity to change the world
- rejects that there is a reality separate from our perception or observation of it (perception is a function of the human mind -> we can only get knowledge of the external world through critical reflection on perception: we need to look into the way we interpret + become aware )= breaks with positivism and scientific realism
17
Q
structure-agency debate
A
- ontological: do agents or structures come first?
- methodological: should explanations of social phenomena be expressed in terms of individuals and relations between them: or can they also invoke social phenomena
18
Q
interpretivism
A
- social world an natural world are fundamentally different: it is subjectively created
- hermeneutic method: rather than seeking to explain/predict by means of laws, the goal is to achieve an understanding of behaviour through interpreting the meanings, beliefs and ideas that give people reasons for acting (as these form behaviour)
- social world doesn’t exist independent of our knowledge of it -> we can’t explain/understand using objective causal laws
- rejects empirical scientific methods: it’s merely brute data (Taylor): looks at behaviour that has a clear physical endstate, not at the meaning of political behaviour for the actor themselves
- emphasizes understanding the meaning that social behaviour has for actors
- see individuals as unique
19
Q
hermeneutics
A
hermeneutic method: rather than seeking to explain/predict by means of laws, the goal is to achieve an understanding of behaviour through interpreting the meanings, beliefs and ideas that give people reasons for acting (as these form behaviour)
- hermeneutics = theories and methods used to interpret texts of all kinds (any object that can be treated as a text (i.e. interpretable). e.g. human beings)
Taylor: any field of study can be the object of hermeneutics if it: contains a field of objects that is a text or text-analogue + is in some way unclear
20
Q
(brute data)
- Taylor
A
data whose validity cannot be questioned y offering another interpretation or reading, data whose credibility cannot be founded or undermined by further reasoning
e.g. voting results: people that raise their hands vote
brute data captures subjective meanings, but not non-subjective meanings such as:
- intersubjective meanings: meanings that do not exist only in the minds of agents, but are rooted in and constitutive of social relations and practices (e.g. paying taxes and voting)
- common meanings: involve recognition or consciousness of shared beliefs, aspirations, goals, values, and common reference point for public life of a society
Taylor: we need to study non-subjective meanings to comprehend political issues + for comparative politics (otherwise study of own categories)
21
Q
political participation positivism vs interpretivism
A
- positivism sees political participation as objective, and often as voting
- interpretivism argues it is subjective and not necessarily (just) voting
22
Q
interpretivism + positivism as two grand traditions
- exagerrated?
A
is exagerrated: don’t differ e.g in methodological conventions:
- clear differentiation of premises and conclusions
- acknowledgement that sampling strategies matter
- recognition that some standards of validation must be established for the sources of evidence used
= differentiation of causes from correlations
etc.
23
Q
internal vs external explanation and evidence
A
- external = often positivist research
correlations or deductions on the basis of ascribed reasons
(evidence based on empirical evidence) - internal = often interpretivist research
concerned with the world of meanings inhabited by the actor and with detaild interpretive work on specific cultures
(based on interpretive evidence)
24
Q
example/case - the analysis of ethnic conflict: a positivist/RC, interpretivist/constructivist and critical realist approach
A
ethnic conflict Yugoslavia 1990s -> collapse of the country
research seeks to find an alternative to the ‘ancient hatred’ explanation, as this seemed wrong: they lived together peacefully for 60 years before a rapid polarization
positivist / Rational Choice = strategic self-interested calculation
- James Fearon: ethnic conflict because SU collapse lead to a commitment problem (lack of powerful third party/state) that can guarantee agreements between the two communities (minority feared their rights would not be protected)
- this follows the hypothetico-deductive method: set of initial determining conditions (no third party + group anticipates that its ability to make or witdraw from agreements will decline + fighting is preferable for this group over the worst scenario that could arise due to the decline) +key mechanisms/general laws (change in relative military power, relative size of ethnic minority, costs for fighting, possibility to establish institutions that protect minorities) (are hypotheses that make ethnic war more or less likely)
- Fearson = generizable over other cases + shows there is a fit between deductions of theory and observed behavioural outcomes
interpretivist / constructivism (social constructed + identity not fixed) = looks at cognitive features (e.g. norms, intersubjective understanding)
- Somer: public discourse forms identity + influences perceptions of ethnic identity. change in dominant picture (ethnic entrepreneurs) -> change in individual behaviour -> change dominant picture = cascade process that is hard to stop
- Somer: public discourse promoted by the state discouraged people from openly expressing their ethnic prejudices (preference falsification: motivating people to publicly falsify their private believe) -> concealed the private importance of the divisive image + there was upward ethnic preference falsification (exaggeration of public support for the divisive picture)
- evidence: survey research + public expression of the divisive picture (public discourse was more intolerant and divisive than private attitudes were)
critical realist = more focus on structure (agents and structure are interconnected)
- VP Gagnon: violent conflict along ethnic cleavages is provoked by elites to create a political context where ethnicity is the only politically relevant identity + such strategies are responses by ruling elites to shifts in the structure of domestic and economic + by constructing individual interest in terms of the threat to the group, endangered elites can fend off domestic challengers who seek to mobilize the population against the status quo, and can better position themselves to deal with future challenges
- structures that provided these conditions for elites: secondary economies (poor functioning planned economy) + constitutional reforms 1974 (Republicanizarion) => corruption
25
critical realist
social phenomena don't exist independently of the agents conceptions of what they are doing in their activity
- structures are the ever-present condition of human agency
- agents interact with and reproduce/transform the structures that enable and constrain their actions
e.g. Morphological Approach (Archer): structural conditioning -> social interaction -> structural elaboration
e.g. retroduction: moving backwards from some initial puzzling fact to the mechanisms that produced it (look what could have accounted for ...)
26
graph page 59 and 60
look at them
27
fact-value distinction
positivism = facts and values are fundamentally different and distinguishable (because: social world is same as natural world (ontology) -> knowledge of observable facts that exist independently of the observers' values = science + fact rather than value)
critical theory = approaches that reject that facts and values can be distinguished: social inquiry is not value-free
- e.g. post-structuralism, constructivism, feminism, post-colonialism
28
(post-structuralism)
- social realities are discursively produced
- cannot be understood independent of their dynamic historical construction in social and cultural discourses
- the way people talk about the world reflects the success of particular ways of thinking and seeing, rather than objective truths
- ways of thinking become invisible because they are assumed to be truthful and right
- purpose of theory to unpack discursive processes (not to establish explanations or seek causality)
29
empirical vs normative theory
distinction shows influence positivism in the field (they say fact/value is distinguishable)
- empirical = hypotheses of 'what is' = objective, value-free
- normative = hypotheses of 'what ought' = values, morals
research practice tends to follow this distinction (looking at facts vs undertaking rigorous, philosophically informed normative study)
*some researchers argue this distinction is wrong: they argue that all theory is normative (there is always normative assumptions, in selecting cases, data etc.): theory is always for some one and for some purpose (Cox)
we must be sensitive to ascribe them as different, we must recognize they are not independent, there are costs to keeping them separate
30
Max Weber on objectivity in social science
+ Nagel
it is difficult to obtain, but possible
1. there is a fact/value distinction
2. value-neutrality is not possible in the social sciences (always selection + interpretation + drawing conclusions)
3. we must strive for a value-neutral social science (by following norms and practice of good social science + by being clear what statements are logically deduced or empirically observed, and which are evaluative)
*Nagel: selection in what to study doesn't have to have impact on the scientific quality/objectivity of the research + social values play a role into the assessment of evidence and conclusions, but that this can be minimalized/eliminated when acknowledged +
31
Rosenthal effect
(researcher bias)
impact the bias of the researcher's expectation on the behaviour of the research subject
e.g. when the expected outcome of the experiment is revealed to the subjects
still: objective social science research is possible: we can minimize contact researcher and his subjects (Michael Martin)
32
research process biases
- Heisenberg effect
- going native = whilst pursuing research on a subject, the researcher's interaction with the subject creates a bias, makes him subjective (e.g. researcher can identify himself with the group)
- behaviour of the subject changes as a result of our observation (e.g. Heisenberg effect: people behave differently when they know they are being observed)
*we can work around this e.g. by using content analysis: looking at past recordings, statements and texts
- behaviour changes as a consequence of the results of a study (so: is the explanation in a study correct, or is it a self-fulfilling prophecy?)
33
Kuhn and fact/value distinction
- observation is theory-laden : observation is the interpretation of a phenomenon in the light of some theory and other background knowledge (+ could even argue that a phenomenon is an already interpreted regularity)
*so not Popper's: theory and observation are independent
- science is a social institution with common views (paradigms), conceptual schemes that defines objects of investigation and the ways of investigating -> science tends towards paradigmatic conformity and conservatism (rather than seeking falsification)
34
Kuhn and paradigms + scientific revolutions
(book: the structure of scientific revolutions)
- growth of scientific knowledge is not a logical proces, but a social one
- science is a social institution in which scientists are **socialized** to accept reigning values, beliefs, concepts, and rules of their profession -> the world appears 'theory-laden'
- paradigms direct 'normal science': science that forces nature into the preformed, inflexible box that the paradigm implies + what doesn't fit in the box isn't seen/studied
- validity of truth-claims is grounded in the consensus of some scholarly community
!challenges fact/value distinction in both social and natural science (!!!!Kuhn considered social sciences to be pre-paradigmatic + unlikely to be capable of paradigm-driven)
history of science = paradigm shifts
model:
1. pre-paradigmatic scientific investigation (random, diverse investigation in lack of a paradigm)
2. establishment of a paradigm
3. normal science (develops/maintains the paradigm)
4. anomaly: something deviates from general rules, basic assumptions of the paradigm
5. crisis (when there are many or significant anomaly(s)) -> loosening of rules as basic institutions are rejected, competing camps (want old vs new paradigm)
6. scientific revolution: new paradigm
= challenge of the positivist ideal of the (rational) scientific process
35
paradigm
- incommensurable
Kuhn: a conceptual scheme, about which some community of scholars agree, and which defines the objects of investigation and the ways in which they are to be investigated
- paradigms are incommensurable: paradigms disagree about what the data are -> paradigm choice can never be settled by logic and experiment alone, it requires some faith
in the study of politics paradigm = world views or schools of thought that provide a theoretical architecture for pursuing the stuidy of the social world (each paradigm has assumptions that provide a basis for the development of theories
36
(example: paradigms and the study of development)
+ Geddes (paradigms + chaos)
+ Higgott (research programmes + sound basis and co-existence)
2 theoretical perspectives/paradigms development studies: modernization theory & dependency theory
modernization theory
- liberal, neoliberal assumptions
- origin: US wanted developing world not to be drawn into the Soviet communist block -> social science to study how to promote capitalism and political stability (1950s)
- assumes underdevelopment is a stage that all nations must pass + that progress comes trough spread of modern capitalism to traditional/backward areas (in which advanced countries had to play a vital role (its not a natural process)
- 1960s: anomaly = modernization didn't appear to happen in developing countries (industrial development appeared to cause worse conditions of life rather than better)
dependency theory
- marxist and neo-marxist assumptions
- capitalist development as a process that delivers unequal benefits
- late 1960s: explained anomaly modernization theory
- argues that international market relations work to reinforce uneven relations developed and developing countries -> developing countries unable to move beyond limited industrialization
- anomaly: demonstrated capacity of some third world countries to industrialize and achieve rapid eco. growth (response: sectors are controlled by multinationals = more loss than gain)
- criticized for: exaggerating power of international system + ignoring the role of internal factors + failing to accout for different developmental outcomes
this debate/struggle -> development studies often seen as not scientific driven, but driven by incommensurable, irreconciliable ideological commitments
Geddes: both paradigms ignore disconfirming evidence + scientists are drawn to a theory more out of their personal beliefs, intuition, emotion than the explanatory abilities of theories
Higgott: they aren't paradigms, they are research programmes with a sound core, they co-exist and challenges don't create chaos and creation of a new paradigm, but changes in the protective belt of the theories
37
post-normal science
Silvio Funtowicz + Jerome Ravetz
an approach that is complementary to, but different from, conventional/normal science
argue:
- normal science appropriate in situations with low levels of uncertainty and risk
- post-normal science when decision stakes or system uncertainties are high
call for a democratization of science: incorporating multiple viewpoints + extending peer community
38
Irme Lakatos and scientific research programmes
! talks about natural sciences + was applied to social sciences
- rejects Kuhn's model of scientific progress (argues it reduces science to mob-psychology (don't pick theories out of consensus, there are rational bases for choosing theory A over B) + that there are always opposing views (so: no normal science))
describes how the incremental, cumulative, progressive articulation of scientific research programs leads to the (critical and rational) growth of scientific knowledge
*there are always competing research programmes + each is developing progressive or degenerative + the acceptance of a theory over an other theory is based on how well it explains novel facts or anomalies (it is based on the degeneration and progression of different research programmes) = rational, progressive process closer and closer to the truth
scientific research programme = hard core + protective belt
- hard core = general hypotheses
- protective belt = auxiliary hypotheses (specific falsifiable hypotheses based on the hard core)
*problem shift = adjustments to the protective belt (can be progressive or degenerative)
negative heuristic = the stipulation that the hard core of the programme not be abandoned or modified
positive heuristic = instructions as to how to adjust the theories and auxiliary hypotheses in the protective belt in the face of an anomaly
!! theory is to be judged on the basis of putting forward novel facts -> PROBLEM: not all facts are equal in significance + it is up to researchers to judge whether programmes are degenerating or are undergoing a temporary lull (can only be judged in retrospective, so choice in the moment is base on subjective valuations)
39
the role of values in research: Kuhn and Lakatos compared
same = scientists work within frameworks of theoretical assumptions + maintain its central set of propositions + Kuhn argues degeneration of a research programme and crisis-stage of a paradigm are the same (Lakatos argues it differs: Kuhn = political and cultural factors explain change, Lakatos = rational process)
different = explanations why scientists maintain a tradition's core assumptions:
- Kuhn: tendency to encourage paradigmatic conformity
- Lakatos: rational methodological decisions
it seems Kuhn describes how scientific practice has been + Lakatos how it should be
