TheoryGreat+Books


 * 3 traditions in CRS**

__Previous questions from comps:__ 1. Briefly outline three major theoretical traditions in the field of comparative and regional studies. What are their similarities and differences? To what extent can they be combined? 2. Briefly present the ontological assumptions of two or three major methodologies currently used in the field of comparative and regional studies. Given these assumptions, to what extent can these different methodological schools be combined and used fruitfully to understand socio-political processes and outcomes? 3. Lichbach and Zudkerman argue that scholars in the field of comparative field of comparative and regional studies employ three types of methodologies: centralist, rationalist, and structuralist. Briefly present the ontological assumptions of these methodologies. Given theses assumptions, to what extend can these different methodological schools be combined and used fruitfully to understand socio-political processes and outcomes. 4. What do theories of postmodernism/poststructuralism contribute to comparative and regional studies? What are their strengths and weaknesses for analyzing concrete instances? Answer with reference to specific authors and empirical referents. 5. What are the principal costs and benefits associated with the use of rational choice theory in the study of comparative politics? 6. What are the principal similarities and differences among rationalist, institutionalist/structuralist and culturalist/discursive methodologies? What are the benefits and pitfalls of using more than one of these methodologies in a single study? 7. To what extent is it (im)possible to integrate the three broad epistemological schools of rationalism, institutionalism and culturalist approaches into a single analysis? Be sure to discuss actual works that exemplify your position. 8. Unlike physics, the study of international relations has multiple methodological approaches to explain socio-political interaction. a) Why are there multiple methodological approaches? b) Pick two and explain their ontologies and epistemologies. c) Is it advisable to use multiple methodological approaches in a single study? Why/Why not? Rational Choice (Aug.30 2010) This is the first of the three ‘competing CRS traditions’ as outlined by L&Z (1997) A NOMOTHETIC theory – interested in discerning covering laws that will apply across people and places. My notes are pretty thin for this week, so I’m adding wiki on rational choice: “__ Rational choice __ theory, also known as choice theory or rational action theory, is a framework for understanding and often formally modeling social and economic behavior. It is the main theoretical paradigm in the currently-dominant school of microeconomics. Rationality ("wanting more rather than less of a good") is widely used as an assumption of the behavior of individuals in microeconomic models and analysis and appears in almost all economics textbook treatments of human decision-making... Rational choice theory uses a specific and narrower definition of "rationality" simply to mean that an individual acts as if balancing costs against benefits to arrive at action that maximizes personal advantage... In rational choice theory, these costs are only extrinsic or external to the individual rather than being intrinsic or internal. That is, strict rational choice theory would not see a criminal's self-punishment by inner feelings of remorse, guilt, or shame as relevant to determining the costs of committing a crime. In general, rational choice theory does not address the role of an individual's sense of morals or ethics in decision-making. Thus, economics Nobelist Amartya Sen sees the model of people who follow rational choice model as "rational fools."... However, it may not be possible to empirically test or falsify the rationality assumption, so that the theory leans heavily toward being a tautology (true by definition) since there is no effort to explain individual goals. Nonetheless, empirical tests can be conducted on some of the results derived from the models. In recent years the theoretical vision of rational choice theory has been subject to more and more doubt by the experimental results of __behavioral economics__. This criticism has encouraged many social scientists to utilize concepts of bounded rationality to replace the "absolute" rationality of rational choice theory: this points to the difficulties of data-processing and decision-making associated with many choices in economics, political science, and sociology. More economists these days are learning from other fields, such as psychology, in order to get a more accurate view of human decision-making than offered by rational choice theory. For example, the behavioral economist and experimental psychologist Daniel Kahneman won the Nobel Memorial Prize in Economic Sciences in 2002 for his work in this field.” As described by Steve: __behavioral economics__ combines psychology and economics. It looks at rational choice ideas like utility maximizing, but then applies psychology to adjust the utility maximizing approach, for example, for future discounting. It is sort of a reaction to rational choice – saying that RC assumptions are flawed. This is different from __behavioralism__: which says that there are behavioral laws that can cover all phenomena (though are not necessarily global in scope). Behavioralism is also not rational choice: - behavioralism starts with data, does analyses, and then builds models. – an empirical approach. EXs: COW dataset/Inglehart on values. - rational choice starts with models and checks data later (but is still a positivist approach). And then there’s the __behavioral assumption__: if there’s a correlation between and IV and DV, there is enough reason to search for causation. Rational choice is very much based on models. There are 2 questions you can ask of a model: (1) How closely does it resemble reality, and (2) what is its predictive power. [I’m pretty sure the latter is the defining question for rational choice, since its often clear that their assumptions may be somewhat unrealistic]. Critique of rational choice: you can set anything as a person’s preferences, as these are generally not empirically verifiable but instead are ‘revealed preferences’. This is circular and can become tautological. Key concepts discussed in class: - Pareto efficiency: nobody can be made better off without anybody else being made worse off. - Kuhnian shift: a paradigm shift that completely displaces the pre-existing approach. Cliché example is the shift from Newton’s to Einstein’s understandings of gravity versus space-time in physics. - Prisoner’s dilemma: cooperation is better for both, but defection is better for each. A key example is nuclear disarmament. =READINGS on rational choice=

Example of rational choice
__Descriptions of rational choice:__ The key factor in rational choice is individual preference orderings which can be aggregated into utility functions to mathematically map out what people will choose, based on their ‘expected utility’. Game theory is a way of modeling how these preferences interact when more than one individual is making a decision at once. Rational choice is instrumental – decisions are evaluated based on their outcomes, not for their own sake. “Intentional and rational actors generate collective outcomes and aggregate behavior that are often socially suboptimal and personally undesirable. This claim... has long been the major contribution of rational choice scholarship” (117). The pillars of the rational choice model are: 1. assumptions of rationality 2. nature of constraints on behavior 3. determinants of strategic interaction, and 4. equilibrium conditions. However, the RC literature has developed, and now encompasses what Levi calls ‘comparative and historical rational choice’ which takes into account more factors, such as context, institutional arrangements, structures, and power. This new work blurs lines between different approaches, is methodologically pluralistic, sefl-conscious about rationality assumptions, and has an increased awareness of the influence of history and contingency. These new theorists are also incorporating behavioralist assumptions and notions of power.
 * Olson** looks at collective action problems and asks under what conditions rational actors would be expected to cooperate to overcome collective action problems. Develops a variety of theories based on group characteristics and different mechanisms that can be used. For example, smaller group members will be inclined to free-ride on larger members of a group, and the larger a group, the less we would expect it to further common interests. Also looks at differences between market groups (want few members to buy in to excludable collective goods) and non-market groups (want many members to help pay for inclusive collective goods).
 * Elster (ch.3)**
 * Levi** –

Critiques of rational choice
=Green and Shapiro= Find that rational choice has failed because it has produced nothing that is both interesting as well as empirically sustainable – all interesting ideas end up being wrong, all ideas that hold up to theory are so simple your grandma could have figured it out. Level two big criticisms at rational choice: 1.) RC is too method-driven (rather than being problem-driven). This means that even the problems that are chosen to be worked on are a priori defined by the methodology. 2.) RC is too universalistic – instead, G&S argue that they should aim for lower level hypotheses and generating middle level theory. This is best done through problem-driven research that seeks answers to real-world puzzles rather than the best theory in the discipline. Rationality can break down for a variety of reasons, including indeterminacy (inadequate evidence, or equal preference rankings). Elster suggests natural and social selection models instead, but notes that these move far too gradually to explain real change in the world. Elster also notes that there are capacities for collective action, and often these are seen as guided by social institutions, organizations and/or bargaining. Concludes that social change is far too complex to be reliably mapped or planned for with any available social theory tools. NICE NOTE FROM ELA – note that Elster’s institutions are seen as much more benign than those proposed by Moe, who sees the role of power as being much more fundamental. =Moe= Whereas rational choice theories see political institutions as structures for voluntary cooperation, Moe says that the structures of institutions also depend heavily on power resources. As a result, while they may be cooperative, they may also be less benign. For example, in democracies, bureaucracies force the losers to comply with the will of the majority. And in the international sphere, big powers can form organizations purely in their own self-interest that smaller powers will be forced to join in order to not lose out completely [NOTE – this circumstance is addressed in cases known as “suasion games”, but I don’t think Moe goes into this].
 * Elster (after Ch.3)**

Other readings
This is the second of the three traditions as outlined by L&Z (1997) A NOMOTHETIC tradition (but also has idiographic tendencies) - Willing to accept some degree of laws, but they are granular and require deep analysis. Generally done with case studies, but keeps an eye to middle range theory. - The key argument is that **context matters**, and the tradition began in response to behavioralists and statisticians who wanted to run a regression on everything without concerning themselves with the background that got people to that juncture in the first place. Starts in the 1980s with March and Olson article ‘The New Institutionalsim: Organizational Factors in Political Life’ - It’s a reaction to behavioralism – says that context matters. You need to look at //how// things happened, not only why. Asks how agendas are set, who holds the power, and how choices are shaped. Steve’s example – its not enough to look at post-highschool decisions of whether or not to go to college – you have to look at the factors shaping those decisions to understand them (background, poverty, etc) In the 1990s, institutionalists get into a big fight with rational choice scholars In the 2000s, a synthesis emerges. An institution is NOT an organization. It’s a constraint on actions or a ‘rule of the game’. A constraint is something like property rights, which are widely maintained but for which there is theoretically some level of choice involved in following them. Economic institutions tend to be quite unstable – they are generally thought to be mutually agreed upon and self-enforcing à EX: a treaty Political institutions are often more stable à EX: welfare state in advanced capitalisms. Historical/rational choice institutions tend to be based on the idea that maintaining the institution increases efficiency, whereas sociological institutions may be maintained for ideals such as politeness or social legitimacy. à EX Sociological institution: stand right/walk left on escalators - This is a common problem in all schools of thought, but comes up especially for institutionalists. Essentially, you have to pick a starting point from which to begin your analysis, and while they can talk about why institutions are important, there’s often little discussion of where they come from. - Theden says there are four types of shifts that can lead to institutional change: - Displacement – e.g. G20 displaces G7   - Drift – e.g. people drifted away from labor unions - Conversion - Exhaustion – arrangements no longer work – e.g. Soviet social planning. - At **critical junctures** there is room for change. Often these are described as dialectic moments in which order becomes disorder and change becomes possible (yep, its true, I STILL don’t get what a dialectic is…) This is largely a critique of behavioralism – it explains decisions that wouldn’t make any sense in a rational choice framework – such as following an institution even when it is not in your interest. - This is similar to a **punctuated equilibrium** model (originally drawn from biology) popularized by Gould. The theory is that institutions in equilibrium can withstand a certain degree of external change before they are suddenly forced to collapse (‘the straw that broke the camel’s back) - There’s also a ‘**learning effect’**: at first an IV won’t affect a DV much, but with learning the relationship becomes more robust. Hall and Taylor (1996) – ‘Political Science and the 3 New Institutionalisms’ Says that there are 3 institutionalisms: - Historical institutionalism - Has a ‘flavor of Marxism’ (Steve’s words) – its inspired by Marx’s ‘18th Brumaire of Louis Bonaparte’ which says that ‘men make their own history, but under circumstances not of their own choosing. - [more in H&T summary] - Rational choice institutionalism: - Includes economic institutionalism o Starts with Coase who asks ‘why do companies exist’, and answers that its because of transaction costs and imperfect information - [more in H&T summary] - Sociological institutionalism - [more in H&T summary] =READINGS on institutionalism=
 * Bates** – this is a key reading for the three traditions questions, but also for methodology questions, as it speaks to levels of understanding. Bates argues that there are three levels of comprehension, but that these are complementary, not contrasting.
 * 1) Apprehension/verstehen – these approaches speak to the particular rather than the ‘system’, and aim for rich understanding. He mentions the ‘new institutionalism’ as part of this, as it bridges this apprehension with more formal theory. The culturalist tradition would also belong though it is not noted explicitly. It is a nomothetic understanding (not Bates’ word)
 * 2) Explanation – as the researcher starts to distance themselves from the field, they aim for explanation – an account that includes ‘therefore’. Some facets of game theory (rational choice) are included under this, namely those that actually try to model real-world events.
 * 3) Conviction – but you’re still not done once you provide a plausible ‘therefore’ account. Now you have to test your explanation against data other than the data from which you derived it, and this can be done through either small- or large-n research.
 * Institutionalism (Sept 13)**
 * What is an institution?**
 * Types of institutions**
 * How to explain change?**
 * - Path dependency** is key: there are only a few ‘switching points’ where its possible to go to a new institution because there are transaction costs of switching paths once you’ve started down one.

About new institutionalism
=Hall and Taylor= Outline three branches of institutionalism (as noted above) =Pierson and Skocpol= Provide a defense of historical institutionalism against rational choice or behavioralism. 1. It asks big substantive real-world questions rather than providing methdod-driven research. 2. It traces historical processes, looking for both causation and description, but seeking causation through complex causal mechanisms and path dependence. 3. Analyzes institutions within their broader contexts – both as they are created and as they evolve. The tradition rejects the ‘functionalist’ (RC) notion that institutions only deal with the issues they were created to solve. Methodology – is often small-n, but with an eye to causal inference so that knowledge can accumulate. However, does not rely on isolating independent variables (as in behavioralism) because it believes context is important. =March and Olsen= Talk about why an earlier article of theirs was cited so much. Say that it was because it recognized a shift in social science from an emphasis on a logic of consequences (like in RC – says behavior is driven by evaluating consequences of actions) to a logic of appropriateness (like in HI – says that behavior is motivated by commitments to identities and rules). They emphasize the need to go beyond in depth case studies to start building generalizable models of institutional change, diffusion of rules, and endogenous generation of new rules.

Examples of institutionalism
=Hacker= Hacker starts with Pierson’s theory that opportunities for retrenchment of welfare programs are highly path dependent, because existing policies create vested interests in maintaining the status quo. Hacker says this is true, but ignores more subtle mechanisms of retrenchment that fall short of full-on policy revision. These alternative mechanisms are (1) policy drift – operational changes without structural changes – i.e. situations on the ground change but welfare organizations are purposefully not updated to cope with them; (2) conversion – when an existing institution is redirected to new purposes; and (3) layering – when new institutions are added without removing the old ones in order to allow change without pissing off vested interests. =North= Talks about institutions, institutional change, and organizations. This is more of a list for what you can say about the authors talking about culture/ideas/identities. If we keep our CRS focus, we are trying to understand states, power relations, modernization, classes etc. The main point in this school of literature is that everything is relative. There is no objective meaning. Meanings are created through rhetoric and communicative action. The first set of readings we had related to this understanding was critical theory (Horkheimer and to an extent Habermas). This critical theory is kind of a mixture between Marxism and postmodernism. You still have the alienation and emancipation concepts, but on the other hand, they are not structurally shaped. They are not bound by the economic relations. The second set of readings was more pure postmodernist (please don’t tell Patrick Jackson I called him a postmodernist). Here, everything is based on social interaction and language. This is why human agent comes up as an important level of analysis. The last set (or the one) was a critic to this understanding. Horkheimer claims the following: - Framework for our study is the interaction if individual activities o The world is a product of the activity of society as a whole. - A theory should fit the needs of people. - The facts are based on the historical character of the object perceived and historical character of the perceiving organ. - But still, Horkheimer keeps his Marxist tradition o Social totality in present is composed of individual and society § Society might put limits on individual behavior. § Critical attitude is still part of this world § And here you have a dialectic identity Habermas claims the following: - Rationality is a construction. - Communicative action is the way to construct everything. o There are some assumptions in the society and this is how you communicate. o But people violate these assumptions, thus systematically distort communication because of  § Intrapsychic defense (this part is kinda Freudian) § Social interaction (this part is Marxist) o Language affects behavior, which in return affects system. o Communication has a hermeneutic and scenic understanding. (Like based on content and context) - ‘Lifeworld’ is based on communication, agreement, and consensus. o Lifeworld is culture, society, personality. o So, it is like the common experiences of people. It is affected by culture, society, and personality. - Cultic practices, religion, communication, and purposive activity messes things up. Foucault claims the following: - Again, everything is based on how you conceptualize it  - He has several problems with ‘normal’ - You have power, if you create knowledge. Also, creating knowledge gives you more power. o Power has two points: § Reorganization of right that invest sovereignty § Coercive forces o The end result is a society of normalization. o Here are the powers for normalization § Hierarchical observation (Panopticism – separate people/observe) § Normalizing Judgment (power – knowledge. Define what normal is) § Examination (Individuals are cases, not citizens) - He takes things to a genealogy level (let’s trace people in the history to see what is going on). Brass claims the following: - Foucault is a great person, he taught us a lot of things. - His(Foucault’s) conceptualization of discursive, genealogy, historical ontology etc. can be used to ask new questions in  o Political theory o Comparative politics o Public administration Jackson claims the following: - Human agent is important in analysis. - Legitimacy is a rhetorical concept that defines the boundaries of acceptable action. Dobbin claims the following: - Cultural relativism is everything. - Everything is based on time and place. Hacking claims the following: - These people are all stupid. - They all assume that social constructs are unnecessarily constructed, they are bad and they should be removed/changed. - They are [insert an academic work of incompetent idiots] because: o There are actually objective bench marks, and facts in the fields they study (So it is not all constructed) o Their categorization understanding is nominalist, it is wrong. o Their external stability understanding is wrong.
 * Constructivist/Critical Theory**
 * When talking about the first set:**
 * Social science methods **

1. KKV assert that there is a single logic of inference that underlies all good quantitative and qualitative research, namely, the scientific rules of inference (scientific method). Do you agree? First, justify your position. Second, give a brief (but fair) synopsis of an opposing view. Third, explain why you think the opposing view is ultimately not convincing. 2. Traditional comparative techniques attempt to generalize at the level of individual cases, seeking covering-laws that can account for systematic relationships between variables. Charles Tilly has argued that comparative work should instead seek to generalize at the level of causal mechanisms that replicate across cases but appear in different configurations. Discuss the strengths and weaknesses of each of these comparative approaches. [this could also be seen as a middle-range theory question]
 * __ Questions from the comp.s __**

=KKV [notes from class]= Main argument: case study researchers need to be more systematic, mainly by borrowing from quantitative methods. Key ideas: - inference: using known facts to create new knowledge (using facts we know to learn something about facts we don’t know) - descriptive inference: using known facts and creating the new knowledge of a relationship between them. For example sun=> warmth – the new knowledge is describing the relationship. - fundamental problem of causal inference: in social science you can never replay an event in a clinical setting, so social science cannot be an experimental science. Suggestions for research: - if you can make more cases out of your data, do it. - variance on both the DV and IV is good, and in small-n research its very important to include this variance consciously. Other key concepts from the book: - causality and the causal effect – the difference between the systematic component of observations made (on the DV) when the IV takes one value versus another. - Assumptions of unit homogeneity and conditional independence - How to judge the unbiasedness versus efficiency of a descriptive or causal inference - Guide to building good causal theories 1. falsifyability 2. internal consistency 3. selecting DVs that are not endogenous and vary across the breadth of cases being explained. 4. maximize concreteness of observable implications 5. state the theory in as encompassing a manner as possible. - leverage – explaining a lot with a little Mahoney (2010), After KKV: The New Methodology of Qualitative Research. [notes from reading] Main argument: KKV was a good start, but there’s a lot more legitimate qualitative methodology that is not based on a regression analysis model. 3 themes in the new research on qualitative methods: (1.) process-tracing and causal process observations - KKV see causal process tracing insufficient – finding intervening variables can help increase the number of observations, but can lead to ‘infinite regress’ as there are potentially unlimited intervening variables. They see it only as a preliminary descriptive tool in exploring a subject area. - Collier, Brady and Seawright (2004) and George and Bennett (2005) see it as much more fundamental. - Specifically, must distinguish between ‘data set observations’ (DSO) (what KKV use) and ‘causal-process observations’ (CPO). - CPOs can be like ‘smoking guns’ and can (dis-)confirm causal inferences – they can be used //both// for theory testing and theory building. - 3 types of theory testing CPOs: (a) independent CPOs: did an IV occur in the manner and/or time posited by the theory. --> EX: did a meteor even hit the earth to wipe out dinosaurs? is there evidence of big bang that could have generated the universe. (b) mechanism CPOs: is a posited intervening event present at the time and place predicted. G&B (2005) say that the number of mechanism CPOs that are present does less to confirm a theory than the likelihood of these CPOs being present if an alternative theory were at work. --> EX: Skocpol (revolutions) observation that vanguard movements usually only form after major revolts have occurred, and thus cannot be generating the revolts. (c) auxiliary outcome CPOs: are separate outcomes that should have occurred if the theory works in the way that it is supposed to. --> EX Luebbert on red-green alliances (2) Set theory and logic – this includes methods such as qualitative comparative analysis, fuzzy-set analysis, set theory, logic, Boolean algebra, and Wittgenstein’s Family Resemblance Approach. - focus on finding the reason mechanisms work, not on the average impact of x on y. - I think these have a lot to do with distinguishing types, and finding necessary and/or sufficient conditions. (3) Combining qualitative and quantitative research (multimethod research) - Using qualitative to support quantitative --> EX: Lieberman’s “nested analysis”. --> EX: Gerring (other reading) on case studies for theory testing versus theory building. - Using quantitative research to support qualitative findings – Lijphart says use qualitative to develop theory and quantitative to test it. - Mahoney says: case study and statistical research are designed to do different things: cases studies ask why certain things happened in certain cases, whereas statistical studies ask what the //average effect// of a variable is. Gerring (2007), The Case Study: What it is and What it does. [notes from reading] Case studies are becoming important again for a variety of reasons: - realization that large-n is still not an experiment - move towards more quasi-experimental work (often case based) - new models of research for case studies - ‘recent marriage of rational choice tools with case study analysis’ - epistemological shifts away from pure positivism. Definitions: population (total scope of proposition) sample (those cases formally analyzed) case study (intensive study of 1 case) case (spatially and temporally bounded phenomenon) observation(s) Differences of opinion between Gerring and Mahoney [my thoughts]: - Both agree that case studies are important and legitimate social science. - But Mahoney does not agree with the idea that case studies are theory-generating whereas large-n studies are theory-testing --> this is why he includes all the causal process mechanisms that are essentially ways of testing big theories with single cases to ensure that certain key mechanisms are at work. - This goes against Gerring’s assertion that case studies are more likely to have Type 1 error (falsely rejecting the null – i.e. finding evidence of a mechanism where there is only chance) whereas cross-case studies are likely to have type 2 errors (failing to reject the null – not finding evidence of a mechanism that’s actually at work, and attributing variation to chance). - I think Mahoney would also object to the idea that case studies merely try to find a causal mechanism (understand the mechanism by which x causes y) rather than causal effects (identifying the magnitude and relative uncertainty of the causal relationship between x and y). I think that this is what Mahoney is driving at with some of his second set of mechanisms (that I don’t understand as well) that talk about sufficient and necessary causes. Though he would agree that case studies are not designed to find the average causal impact of x on y, I think he would object to the idea that they can do more than describe a casual process. In short, if we had to line these authors up on a scale of least to most enthusiastic about the capacity of case-study research, they would be: KKV, Gerring, and then Mahoney. Elster (Ch.1) (This is notes from readings from week II on rational choice) Says there are both events and facts in the world, and to explain an event is to give an account of why it happened, generally by citing earlier events and facts. Talks about **causal explanations:** - must include causal mechanisms - must talk about more than just correlation - causal explanations are neither necessary conditions, nor are their story telling or predictions.
 * __Notes and info from readings__**
 * || Case study || Cross-case study ||
 * Research Goals ||  ||   ||
 * 1. hypothesis || Generating || Testing ||
 * 2. validity || Internal || External ||
 * 3. causal insight || Mechanisms || Effects ||
 * 4. scope of proposition || Deep || Broad ||
 * Empirical factors ||  ||   ||
 * 1. population of cases || Heterogeneous || Homogenous ||
 * 2. causal strength || Strong || Weak ||
 * 3. useful variation || Rare || Common ||
 * 4. data availability || Concentrated || Dispersed ||


 * Stuff on middle-range theory and nomothetic versus idiographic**

[Note – Efe and Kate are thinking that these exact questions probably won’t come up, since the class didn’t focus on middle-range theory per se. However, it seems to be an important theme, and a good thing to get a handle on so you can toss it in with the other methodology stuff. I’ve included any bits an bobs on it that I’ve come across here] 1. Scholars in the field of comparative and regional studies rank among the most prolific producers of “middle range theory” in the social sciences. a) what is middle range theory? b) what are the strengths and weaknesses of middle range theory in scholarly inquiry? c) have comparative and regional studies in practice delivered in producing the hoped for outcomes of the proponents of middle range theory? 2. Have scholars in the field of comparative and regional studies actually succeeded in advancing our understanding of socio-political phenomena by undertaking “middle range research and aggregating the results? If so, please provide examples of some “settled questions and how they became so. If not please explain why not.
 * __ Questions off the comps __**

Readings and notes on the subject
[From class]: - nomothetic: looks for a covering law, or a general rule. This is the approach taken by KKV, who look for universal laws of social interaction. - idiographic: searches for a deeper understanding, and claims the purpose of research is to get deeper into the cultural story. EX: Clifford Geerts says observable implications are not enough – knowing a wink from a twitch requires social understanding. [From wikipedia]: Nomothetic and idiographic are terms used by Kantian philosopher Wilhelm Windelband to describe two distinct approaches to knowledge, each one corresponding to a different intellectual tendency, and each one corresponding to a different branch of academe. Nomothetic is based on what Kant described as a tendency to generalize, and is expressed in the natural sciences. It describes the effort to derive laws that explain objective phenomena. Idiographic is based on what Kant described as a tendency to specify, and is expressed in the humanities. It describes the effort to understand the meaning of contingent, accidental, and often subjective phenomena.
 * nomothetic vs. ideographic approaches:**

3. Conviction – but you’re still not done once you provide a plausible ‘therefore’ account. Now you have to test your explanation against data other than the data from which you derived it, and this can be done through either small- or large-n research. Case studies are becoming important again for a variety of reasons: - realization that large-n is still not an experiment - move towards more quasi-experimental work (often case based) - new models of research for case studies - ‘recent marriage of rational choice tools with case study analysis’ - epistemological shifts away from pure positivism. Definitions: population (total scope of proposition) sample (those cases formally analyzed) case study (intensive study of 1 case) case (spatially and temporally bounded phenomenon) observation(s) =Green and Shapiro= Level two big criticisms at rational choice: 1.) RC is too method-driven (rather than being problem-driven). This means that even the problems that are chosen to be worked on are a priori defined by the methodology. 2.) RC is too universalistic – instead, G&S argue that they should aim for lower level hypotheses and generating **middle level theory**. This is best done through problem-driven research that seeks answers to real-world puzzles rather than the best theory in the discipline. Key organizational principles of CRS and of the first 4 weeks of class are based on **Lichbach and Zuckerman (1997)**, who claim that there are three competing traditions in comparative politics: These authors address both the idiographic/nomothetic divide, and the 3 traditions, and the middle-range theory. This is based largely on L&Z, but also includes the views of other authors:
 * On middle range theory, the discussion gets a bit muddled up with the 3-theories questions.**
 * Bates** – this is a key reading for the three traditions questions as it speaks to levels of understanding. Bates argues that there are three levels of comprehension, but that these are complementary, not contrasting.
 * 1) Apprehension/verstehen – these approaches speak to the particular rather than the ‘system’, and aim for rich understanding. He mentions the ‘new institutionalism’ as part of this, as it bridges this apprehension with more formal theory. The culturalist tradition would also belong though it is not noted explicitly. It is a nomothetic understanding (not Bates’ word)
 * 2) Explanation – as the researcher starts to distance themselves from the field, they aim for explanation – an account that includes ‘therefore’. Some facets of game theory (rational choice) are included under this, namely those that actually try to model real-world events.
 * Gerring (2007)**, The Case Study: What it is and What it does. [notes from reading]
 * || Case study || Cross-case study ||
 * Research Goals ||  ||   ||
 * 1. hypothesis || Generating || Testing ||
 * 2. validity || Internal || External ||
 * 3. causal insight || Mechanisms || Effects ||
 * 4. scope of proposition || Deep || Broad ||
 * Empirical factors ||  ||   ||
 * 1. population of cases || Heterogeneous || Homogenous ||
 * 2. causal strength || Strong || Weak ||
 * 3. useful variation || Rare || Common ||
 * 4. data availability || Concentrated || Dispersed ||
 * || Rational Choice || Culturalist || Structuralist (/institutionalist) ||
 * General || Starts with the individual and then moves outwards, often using mathematic principles. || Looks at single cases and seeks a rich understanding, but doubts the capacity to generalize from this || Focuses on institutions and histories ||
 * Methodology || Looks for universal laws (nomothetic) but also does purely theoretical work || Highlights reliability over generalizability (idiographic) || Looks for universal theories with causal accounts, but is also concerned with opening up the specifics of each case (both idiographic and nomothetic) ||
 * Ontology (what are the things it studies) || Reason || Rules || Relations ||
 * Explanatory strategy || Comparative static experiments || Interpretive understandings || Historical dynamics ||