Citizen Paine

VIDEO: Howard Dean Tells Dems to Kill Senate Health-Care Bill by AlterNet Staff, AlterNet.

With no public option and no Medicare buy-in, the Senate bill is not worth voting for, the former DNC chairman tells “Countdown.” Former Democratic National Committee Chairman Howard Dean would rather see no health-care bill than a bad one. So he tells MSNBC’s Lawrence O’Donnell.

Are Americans a Broken People? Why We’ve Stopped Fighting Back Against the Forces of Oppression By Bruce E. Levine, AlterNet.

A psychologist asks: Have consumerism, suburbanization and a malevolent corporate-government partnership so beaten us down that we no longer have the will to save ourselves?

Can people become so broken that truths of how they are being screwed do not “set them free” but instead further demoralize them? Has such demoralization happened in the United States? Do some totalitarians actually want us to hear how we have been screwed because they know that humiliating passivity in the face of obvious oppression will demoralize us even further? What forces have created a demoralized, passive, discouraged U.S. population? Can anything be done to turn this around?

Yes. It is called the “abuse syndrome.” Abusive pimps, spouses, bosses, corporations, and governments stay in control [by shoving] lies, emotional and physical abuses, and injustices in their victims’ faces, and when victims are afraid to exit from these relationships, they get weaker.

Does knowing the truth of their abuse set people free [from] abuse syndromes?

No. The truth of their passive submission to humiliating oppression is more than embarrassing; it can feel shameful — and there is nothing more painful. It is not likely that the truth of humiliating oppression [will] energize constructive actions.

Has such demoralization happened in the U.S.?

In the United States, 47 million people are without health insurance, many millions more are underinsured or a job layoff away from losing coverage. Despite the sellout by their elected officials to the insurance industry, there is no outpouring of millions of U.S. citizens protesting this betrayal.  And, the majority of Americans oppose U.S. wars in Afghanistan and Iraq, and the taxpayer bailout of the financial industry, yet only a handful has protested.

[In] the 2000 U.S. presidential election the Florida Supreme Court’s order for a recount of the disputed Florida vote was overruled by the U.S. Supreme Court in a politicized 5-4 decision.  Justice John Paul Stevens remarked: “…the identity of the [loser] of this year’s presidential election…is perfectly clear. It is the nation’s confidence in the judge as an impartial guardian of the rule of law.” Even this provoked few demonstrators.

When people become broken, they cannot act on truths of injustice. Furthermore, …truths about how they have been victimized can lead to shame about [allowing] it, …[and make them] even more psychologically broken.

U.S. citizens do not actively protest obvious injustices [because]…they feel helpless to effect change. The more we don’t act, the weaker we get [and]… move to shut-down mode and escape strategies such as depression, substance abuse, …which further keep us from acting. This is the vicious cycle of all abuse syndromes.

Do some totalitarians actually want us to hear how we have been screwed because they know that humiliating passivity in the face of obvious oppression will demoralize us even further?

Maybe.

Shortly before the 2000 U.S. presidential election, George W. Bush [joked] to a wealthy group, “What a crowd tonight: the haves and the haves-more. Some people call you the elite; I call you my base.” Yet, …citizens who had come to despise Bush and his arrogance remained passive in the face of the 2000 non-democratic presidential elections.  Perhaps the “political genius” of the Bush-Cheney regime was in their full realization that Americans were so broken that the regime could get away with damn near anything… [Even slamming] a boot on their faces.

What forces have created a demoralized, passive, discouraged U.S. population?

The U.S. government-corporate partnership has used its share of guns and terror to break Native Americans, labor union organizers, and other dissidents and activists. But today, most U.S. citizens are broken by financial fears.

The U.S. population is increasingly broken by the social isolation created by corporate-governmental policies. A 2006 American Sociological Review study (“Social Isolation in America: Changes in Core Discussion Networks over Two Decades”) reported that, in 2004, 25 percent of Americans did not have a single confidant. Social connectedness is disappearing in virtually every aspect of U.S. life. There has been a significant decrease in face-to-face contact with neighbors and friends due to suburbanization, commuting, electronic entertainment, time and money pressures and other variables created by governmental-corporate policies. Union and other ways that people support each other to resist oppression also decreased.

We are also broken by a corporate-government partnership that has [taken] control [of] basic necessities of life, including our food supply. We are broken by socializing institutions that alienate us from our basic humanity. A few examples:

Schools and Universities: Do most schools teach young people to be action-oriented — or to be passive? Do most schools teach young people that they can affect their surroundings — or not to bother? Do schools provide examples of democratic institutions — or examples of authoritarian ones?  School is nothing less than a miniature society: what young people experience in schools is the chief means of creating our future society. Kids learn to comply with authorities for which they often have no respect, and to regurgitate material they often find meaningless. These are great ways of breaking someone.

Mental Health Institutions: Aldous Huxley predicted today’s pharmaceutical society “[I]t seems to me perfectly in the cards,” he said, “that there will be within the next generation or so a pharmacological method of making people love their servitude.”  Today, increasing numbers of people in the U.S. who do not comply with authority are being diagnosed with mental illnesses and medicated with psychiatric drugs that make them less pained about their boredom, resentments, and other negative emotions, thus rendering them more compliant and manageable.

Oppositional defiant disorder (ODD) is an increasingly popular diagnosis for children and teenagers [who] “often actively defy or refuse to comply with adult requests or rules,” and “often argue with adults.” A more common reaction to oppressive authorities is passive defiance – e.g., attention deficit hyperactivity disorder (ADHD). Virtually all children diagnosed with ADHD will pay attention to activities that they actually enjoy or have chosen. The “disease” goes away when ADHD-labeled kids are having a good time and in control.

When human beings feel too terrified and broken, they may stage a “passive-aggressive revolution” by getting depressed, staying drunk, and not doing anything — one reason why the Soviet empire crumbled. But, diseasing or medicalizing rebellion and drug “treatments” even weaken this power.

Television: In Four Arguments for the Elimination of Television (1978), Jerry Mander compiled a list of the “Eight Ideal Conditions for the Flowering of Autocracy,” claiming that television helps create all eight conditions for breaking a population.

(1)   Occupies people so that they don’t know themselves — and what a human being is;

(2)   Separates people from one another;

(3)   Creates sensory deprivation;

(4)   Occupies the mind and fills the brain with prearranged experience and thought;

(5)   Encourages drug use to dampen dissatisfaction (while TV itself produces a drug-like effect, this was compounded in 1997 the U.S. Food and Drug Administration relaxing the rules of prescription-drug advertising);

(6)   Centralizes knowledge and information;

(7)   Eliminates or “museumize” other cultures to eliminate comparisons; and

(8)   Redefines happiness and the meaning of life.

Commercialism of Damn Near Everything: Gross commercialization of spirituality, music, and cinema deadens their capacity to energize rebellion. So now, damn near everything – not just religion – is an “opiate of the masses.”

The primary societal role of U.S. citizens is no longer “citizen” but “consumer.” Citizens know that buying and selling within community strengthens that community and that this strengthens democracy, consumers care only about the best deal. Citizens understand that dependency on an impersonal creditor is a kind of slavery, consumers get excited with credit cards with a temporarily low APR.

Consumerism breaks people by devaluing human connectedness, socializing self-absorption, obliterating self-reliance, alienating people from normal human emotional reactions, and by selling the idea that purchased products — not themselves and their community — are their salvation.

Can anything be done to turn this around?

When people get caught up in humiliating abuse syndromes, more truths about their oppressive humiliations don’t set them free. What sets them free is morale.

What gives people morale? Encouragement. Small victories. Models of courageous behaviors. Anything that helps them break the vicious cycle of pain, shut down, immobilization, shame over immobilization, more pain, and more shut down.

The last people to turn to are mental health professionals. Specifically required talents are a fearlessness around image, spontaneity, and definitely anti-authoritarianism, which are not traits medical or graduate schools encourage.

If you want to feel hopeless, there are a lot of things you could feel hopeless about. If you act on that assumption, then you’re guaranteeing that’ll happen. If you act on the assumption that things can change, maybe they will. The only rational choice, given those alternatives, is to forget pessimism.

A major component of the craft of maintaining morale is not taking the advertised reality too seriously.

An elitist assumption is that people don’t change because they are either ignorant of their problems or ignorant of solutions. An elitist who has never been broken by his or her circumstances does not know that people who have become demoralized do not need analyses and pontifications. They need a shot of morale.  READ MORE:

http://www.alternet.org/politics/144529/are_americans_a_broken_people_why_we%27ve_stopped_fighting_back_against_the_forces_of_oppression

SPECIAL BONUS: A Global Philosophy for Successful Living in Eight Aphorisms.

From the BUDDHA: Go forth in joyful participation in the sorrows of the world.

From JOSEPH CAMPBELL: Follow your bliss.

From CHRISTIAN TRADITION: Practice the Golden Rule.

From GHANDI: Act. “Without action there is no result. You may not see the result in your lifetime, but if you do not act, there will be no result at all.”

From JACQUES COUSTEAU: Hope for the best. “I hope for the best, although I can’t say why.”

From TOM PAINE: Use Common Sense. “Reason is the most reliable path to the truth.”

From his holiness the 14th DALI LAMA: “If you want the best idea of how the world was created, don’t pick the best mythology, consult the best science.”

—————————
From FatLemon: “Keep on keepin’ on, and don’t forget to salute the man in the moon.”

LAST THOUGHT:  Don’t get mad, get even.  Fight harder.  Take back the Democratic Party and elect Progressives to all offices in the land.  Continue to fight the oppressive fascist powers.  Bill Hart stood for courtesy, courage, and justice.

For North history matters not only because we can learn from the past, but also “because the present and the future are connected to the past by the continuity of a society’s institutions.” The book outlines a theory of institutions and institutional change. North’s focus is on the problem of cooperation, which allows economies to capture the benefits of trade identified by Adam Smith and his followers. Institutions can evolve to create a hospitable environment for cooperative solutions, but they may also induce economic stagnation and decline. (p. vii) Institutions can, in other words, be efficient or inefficient, and the divergence between institution can explain the different developmental paths found between societies. (p. 6-7)

John Zysman has noted that “rational choice institutionalists start with individuals and and ask where institutions come from, whereas historical institutionalists start with institutions and ask how they affect individuals’ behaviour.” Kathleen Thelen places North in the rational choice tradition, but argues that he embraces a “non-functionalist, more historical view of institutions.” (Thelen, pp. 379-380) North argues that institutions create increasing returns (p. 95), a finding that had an important contribution to to theories of path dependence. (Pierson, p. 256)

North defines institutions as “the rules of the game in a society or, more formally, are the humanly devised constraints that shape human action.” (p. 1) The first part of the book examines the characteristics of institutions and the impact that institutions have on economic (or societal) performance. The second part proposes a theory of institutional change. The third part seeks to explain the differential performance of economies through time.

There is an extensive literature in game theory about the problem of cooperation. Olson’s free rider dilemma showed how cooperation is difficult to sustain when a game is not repeated. Axelrod’s The Evolution of Cooperation was more optimistic, showing how cooperative solutions could be devised in games with multiple iterations. Game theory shows, through varying payoffs, when cooperation can be expected. However, assumptions made in game theory about wealth maximizing individuals who possess equal skill (and access to relevant information) are problematic. Crucially, game theory fails to provide us with a theory of the underlying costs of transacting (pp. 15-16).

North divides his analysis between informal constraints and formal constraints. Informal constraints come from socially transmitted information, and a are commonly referred to as parts of culture. (p. 37) These “self-imposed codes of behaviour” can constrain maximizing behaviour. (p. 43) Cultural processing of information has an impact on the way institutions evolve, and is therefore a source of path dependence. Informal constraints are not merely the extensions of formal rules, and will not change immediately in response to changes in formal rules. (pp. 44-45)

Formal rules include political (and judicial) rules, economic rules, and contracts. They can modify, revise or replace informal institutions in some circumstances. Rules are often devised with private (rather than social) benefits in mind, so the actual structure of rules will reflect the relative bargaining power of different parties. (p. 47) As a result, inefficient property rights that protect the interests of a small group can emerge. These inefficient institutions will not necessarily be eliminated over time. The efficiency of the political market is key, high transaction costs and the subjective perceptions of actors have resulted in property rights that do not encourage economic growth, and actors may have no incentive to create more productive rules. This inertia may be the result of the high transaction costs associated with political change, or a lack of credible commitment that new institutions will be more efficient. (p. 52)

Enforcement is a transaction cost. With complete information, cooperation can be achieved without enforcement. However, with incomplete information, cooperation will break down unless institutions exist to police deviations. These institutions provide information (a communication mechanism) and enforce punishment (as a public good) when cooperation breaks down. (p. 57) A lack of adequate enforcement mechanisms is the most important source of stagnation and underdevelopment in the Third World. (p. 54)

Organizations were defined in part one as “groups of individuals bound by some common purpose to achieve objectives.” (p. 5) In part two North returns to the relationship between organizations and institutions to explain institutional change. He argues that “[o]rganizations and their entrepreneurs engage in purposive activity and in that role are the agents of, and shape the direction of, institutional change.” (p. 73) Institutions can be understood as the rules of the game (or a social incentive structure), and organizations as the players. Neoclassical economics did not expect that inefficient institutions would survive, but in reality such institutions are common. Why is this the case?

Institutions may not (in the absence of perfect knowledge/certainty) reach a neoclassical allocative efficiency, but instead may settle for an a type of adaptive efficiency which allows for a maximum of choices under uncertainty. (p. 80) This type of institutional arrangement may be stable. North argues that institutional stability is achieved “by a complex set of constraints that include formal rules nested in a hierarchy, where each level is more costly to change that the previous one. They also include informal constraints which are extensions, elaborations, and qualifications of rules…” (p. 83)

Fundamental changes in relative prices are the most important source of institutional change. Sources of change in relative prices include changes in the ratio of factor prices (i.e. ratio of land to labour, labour to capital, or capital to land). Change in prices alter perceived costs and benefits of new bargains and contracts. Changes in bargaining power lead to efforts to restructure contracts. (p. 84)

Most institutional change is incremental. However, war, revolution, conquest, and natural disasters are all causes of discontinuous change. Discontinuous change refers to sudden or violent changes in formal rules. However, there is no corresponding change in informal rules. Over time this leads to a restructuring of overall constrains (in both directions) to produce a less revolutionary equilibrium. (pp. 89-91)

Increasing returns and imperfect markets determine the path of institutional change. North gives the example of the Northwest ordinance as an institution that provided adaptively efficient economic development—a fee-simple ownership of land and clear system of inheritance made land transferable with minimum transaction costs. In contrast, unproductive paths explain the relatively poor economic performance of Latin America. (pp. 95-100)

North begins part three by looking at what changes must be made to neoclassical theory to incorporate it with institutional analysis. Next, it outlines the implication for static analysis of economic performance. Finally, he explores a theory of long-run economic change. (p. 107)

North uses the idea of path dependence introduced in the previous section to compare the divergent performance of England (North America) and Spain (Latin America) in the New World. In North America, an institutional framework evolved that permitted the exchange of complex information in a way that captured potential gains from modern technology. In the latter, development was stunted by an institutional framework that privileges personalistic relationships. (pp. 113-117) North’s comparison between North American and Latin America offers an alternative to teleological modernization theory—rather than assuming a final destination a priori, North is interested in “what kinds of institutions are necessary to enable the cost of transacting and transforming to be at the level that will permit this increasing specialization and division of labour to occur.” (p. 120)

North made a vital contribution to our understanding of how institutions affect development. However, because he privileged institutions, he was particularly critical of dependency theory as an explanation for economic underdevelopment. For North dependency theory “not only rationalizes the structure of Latin American economies, but also contains policy implications that would reinforce the existing institutional framework.” (pp. 99-100) Third World countries are poor, not because of post-colonial dependency on the West, but “becuase the institutional constraints define a set of payoffs to political/economic activity that do not encourage productive activity.” However institutional frameworks are not necessarily incompatible with dependency theory. In an increasingly integrated world, institutions cross national boundaries, a point North’s nation-centred analysis does not acknowledge.

A second possible criticism is that North’s institutional approach is, in the words of Stefano Fiori (p. 1025), “not utilizable”. North shows how institutions affect economic performance, but with regard to the question of how to increase institutional efficiency, North only offers a few conjectures. However, North’s focus is historical. Utilizable prescriptions for increasing insitutional efficiency will have to be found elsewhere.

Reference:
Kathleen Thelen, ‘Historical Institutionalism in Comparative Politics’, Annual Review of Political Science, 1999, pp. 369-404
Paul Pierson, ‘Increasing Returns, Path Dependence, and the Study of Politics’, The American Political Science Review, Vol. 94, No. 2, 2000, pp. 251-267
Stefano Fiori, ‘Alternative Visions of Change in Douglass North’s New Institutionalism’, Journal of Economic Issues, Vol. 36, No. 4, 2002, pp. 1025-1043

Rational choice theory has become dominant in political science, but it comes under attack from Green and Shapiro for a lack of “empirical power.” A large part of theory produced by rational choice has not been tested empirically, and even where empirical tests are carried out they are chastised as “banal”—that is they “do little more than restate existing knowledge in rational choice terminology.” The authors argue that “the weaknesses of rational choice scholarship are rooted in the characteristic aspiration of rational choice theorists to come up with universal theory of politics.” As a result, theorists “pursue ever more subtle forms of theory elaboration, with little attention to how these theories might be operationalized and tested.” This leads to the “debilitating syndrome in which theories are elaborated and modified in order to save their universal character, rather than by reference to the requirements of viable empirical testing.” (p. 6)

The critique is focused on studies of American politics, with a particular emphasis on collective action, legislative policy making, and party competition. This literature grew out of three classic texts: Kenneth Arrow’s Social Choice and Individual Values (1951), Anthony Downs’s An Economic Theory of Democracy (1957), and Mancur Olson’s The Logic of Collective Action (1967). Arrows “impossibility theorem” showed that preferences could not be aggregated, and that no social welfare function can exist unless it is imposed. Downs argued that the ideological position of two parties competing for support will tend to converge on the “median voter”. Downs also saw voter turnout as a collective action problem, and argued that it is irrational for individual voters to expend resources gathering information about politics. Olsen expanded the problems of collective action to interest groups. He argued that “only a separate and ’selective’ incentive will stimulate a rational individual in a latent group to act in a group-orientated way.” (pp. 7-9)

Chapter two begins by introducing the main assumptions behind rational choice theory: utility maximization, consistency, each individual maximizes expected payoff, maximizing agents are individuals, and rational choice models apply equally to each person under study (homogeneity). (pp. 14-17) However, there are various differences between rational choice theories. One key division concerns “thin-rational” accounts from “thick-rational” accounts. Thick rational accounts add some description to preferences or beliefs (for example utilitarianism and classical economics). Thin rational accounts do not specify the content of the preferences. They then become tautological by saying that any choice (as long as it meets the requirements of Arrovian weak ordering) is rational. In practice, empirical applications seldom approximate weak rationalism. Rational choice accounts also differ in the amount of information assumed to be available. Neoclassical economics assumes perfect information, but this is unrealistic in politics (voters are often castigated as ignorant). Rational choice in political science in practice makes a range of assumptions about the amount of information actors posses. (pp. 17-19)

Most rational choice accounts share two basic types of explanation. One concerns the existence of intentions as causes. However, demonstrating the existence and causal efficacy of intentions is difficult. Rational choice theorists can be divided between internalists and externalists. Internalists argue for intentional accounts of political action. Externalists argue that we should simply proceed as if intentional accounts are true and draw conclusions about its causal efficacy empirically. Evolutionary biology is externalist because it is not necessarily the product of the intentional states of organisms, and in the same way rational choice can be understood as a model of “powerful selection mechanisms”. However, “external” readings of the social sciences are compatible with a wide range of hypotheses and therefore difficult to test empirically. (pp. 20-23) The second concerns universality. For many rational choice theorists, the search for universal theory is a search for equilibria. The concept of rational choice equilibrium can be traced back to the Nash equilibrium expounded by John Nash in 1950. However, William H. Riker argued that determinate predictions cannot be derived from the laws of equilibrium models—calling political science “the dismal science”. This has led theorists to abandon “all-or-nothing” universalism, and put forward theories of partial universalism (rationality can only explain a part, and not all of what goes on in politics—for example John Ferejohn’s “folk theorem”) and segmented rationalism (rational choice is only applicable in certain circumstances). Others look at rational choice as a family of theories, assuming that different actors will want to maximize different things, depending on the situation they find themselves in (pp. 23-30).

Rational choice theory aspires to be (at least partially) universal. However, much rational choice research is based on unrealistic assumptions (for example people always act rationally, they base actions on certain types of information, the update their beliefs in accordance with Bayes’ Rule, they evaluate options based on the values specified in the theory, the relevant political “commodities” are homogeneous and infinitely divisible, and that preferences remain fixed). How can this be justified? One possible way is Friedman’s instrumentalist approach—a theory is tested by its predictive or explanatory power, not its internal coherence. The authors reject this—either a theory is justified on covering-law grounds (this refers to the fact that scientific advance only comes with developing theory, in which case it cannot be based on unrealistic assumptions), or it is justified on instrumentalist grounds (in which case the mode of theory building is irrelevant, what matters is generating testable hypotheses) (pp. 30-32).

Chapter three looks at methodological “pathologies”, which the authors argue derive from a method-driven rather than problem-driven approach to research. This style of theorizing searches for universalist explanations for politics, and therefore tends towards post-hoc explanations—that is assumptions are designed to fit the data. Besides, because what being a rational actor means if often not clearly specified, it is not obvious what kind of behaviour is (in principle) not explainable by some variant of rational choice theory. And even if rational choice can explain certain phenomenon, it is not clear that other theories could not have equal explanatory validity. (pp. 34-38)

The authors identify problems in formulating tests for rational choice predictions. Often models are so parsimonious or abstract that “all recognizable features of politics are absent.” Others describe their models as general truths that may not always coincide with actual, observed cases which are influenced by various conditions outside the theory. Rational choice models also typically concern many unobservable terms, causing the theory to outstrip the ability of the data to test. Games may be “nested” following Tsebelis, but this is difficult to test empirically. How do we identify which other games actors participate in? A further question concerns how hypotheses should be tested—if millions of people make small donations to a referendum campaign, what does this tell us about Olson’s collective action problem. (pp. 38-42) There are three potential problems with the way rational choice hypotheses are tested. These are biased selection of confirming evidence, projection of evidence from theory, and arbitrary domain restriction. (pp. 42-46)

Voting is an obvious paradox for rational choice theory. Why do people bother to vote when one vote has an infinitesimal chance of altering an election outcome? Voting should be a collective action problem, and yet, each election millions go to the polls. Voting is a failure for rational choice theory, but interesting for Green and Shapiro because it shows the ways that rational choice theorists try to deal with discrepancies between observation and theory.

A decision theoretic model expects that people go to the polls if:
pB+D>C
where p=chance of casting decisive vote, B=benefit of preferred candidate wins, D=utility one receives as a direct result of voting (selective incentives), C=cost of voting.
The obvious conclusion is that the equilibrium level of voting is near zero. So how does rational choice explain voting?
(1) People are civic-minded—this merely substitutes paradox of civic-minded participation for paradox of voter turnout.
(2) Voting is an act of consumption (D)—for example a psychic benefit for doing ones civic duty, or side-payments made by politicians to supporters. However assuming (without evidence) that the selective benefits of voting are higher than the costs is little more than a tautology. It also does not explain why there is not comparable enthusiasm for other forms of civic duty.
(3) The costs of voting are low (C)—Olson argues voting costs are “insignificant and imperceptible” to many voters. However the probability that one vote will alter an election is so low, that even very low costs are hard to overcome. Therefore some argue..
(4) p is not as small as some suggest—people may overestimate the chance of casting the decisive vote, but there is no empirical evidence for this. A single vote may also contribute to a party’s mandate, or may have a non-negligible chance of affecting a vote within a precinct. To assess the likelihood of this, we can only rely on intuition. (pp. 50-56).

Can game theory offer a better account? Theorists suggests that instead of taking p as given, we determine it endogenously by the interaction of strategically minded voters, each facing a similar decision. Initial accounts suggested that this approach could pay dividends, but it collapses once we allow for the possibility that voters are uncertain about voting costs of other citizens or lack information about the precise level of support for competing candidates. (pp. 56-58)

Some have sought to characterize turnout as a “low cost, low benefit” affair that falls outside the boundaries of rational choice. However, this is in “arbitrary domain restriction” for two reasons. First, there is nothing in rational choice theory that specifies when costs and benefits are too low to make the theory inapplicable. Second, voter turnout is not always a “low cost” activity (people turnout despite long lines at the polls or voter intimidation). (pp. 58-59)

As selective benefits increase, statistically people are more likely to vote. However, this does not make voting rational. A marginal increase in utility will increase turnout, even when the costs still outweigh the benefits. Similarly, a greater sense of civic duty, or a sense that the election is close, will increase turnout without making the choice to vote a rational one. (pp. 59-65)

Selective costs of voting (such as poll taxes, long queues) should depress turnout, likewise selective benefits (buying influence with officials, currying favour with friends) might increase turnout. However, there is little empirical evidence to back up the claim that selective benefits increase turnout. The data does support the notion that people vote in greater rates where they think they have an obligation to do so, or want to reaffirm their partisan identity. People feel they have a civil duty to vote, regardless of expected utility (pp. 65-68).

Chapter five looks at the free riding problem in rational choice scholarship. The chapter focuses on how rational choice theory explains voluntary political behaviour other than voting, primarily Olson’s work on collective action.

Collective action problems often take the form of the prisoners’ dilemma game with multiple players. The model assumes that players derive no utility “from doing the right thing”, and therefore predict that no players will adopt a cooperative strategy. Introducing (positive or negative) side-payments may however change the character of the game. One interesting implication of this is that participation in an interest group becomes incidental to the collective good being pursued—a fundamentalist Christian might join a pro-choice rally over a pro-life rally if the sandwiches tasted better! (pp. 74-79)

It may be possible to explain political apathy from the collective action problem, but there are a host of other competing accounts, including a distaste for politics, principled refusal, indifference and so on. Olson uses unorganized (latent) groups (for example migrant farm workers, taxpayers, and consumers) as evidence for his theory. However, it is unclear if these groups are actually agreed about what their interests are, or what a lack or organization tells us about free-riding as opposed to “apathy, ambivalence, or antipathy” towards politics. And of course, we do find examples of mass participation in various groups. What level of participation is compatible with Olson’s theory? Significantly, many studies of collective action provide no control group where comparable benefits are provided directly through individual action—people also tend not to contact the government to resolve personal problems. (pp. 79-83)

Olson recognized the importance of control groups in Logic of Collective Action, and looked at two sets of comparisons: large groups versus small groups and groups that offer selective incentives against those that do not. The claim that selective incentives matter seems intuitively compelling, but the assertion that selective benefits are not is counter-intuitive—indeed empirical studies have found higher rates of participation among those with more at stake in collective outcomes. (pp. 83-85) Revisionist accounts of Olson’s work have tried to get around the problem by redefining what is meant by selective incentives, but in doing so “have robbed Olson’s theory of the predictions that make it provocative and testable.” (pp. 85-88)

Experimental evidence (from laboratory games) only finds mixed support for rational choice theory. Games that are similar in mathematical form, and thus expected to produce similar outcomes, in fact produced levels of cooperation that varied considerably according to the social context in which the game was played. (pp. 88-93) Rational choice theorists could reject evidence from the laboratory as “artificial”, but is this consistent with maintaining that rational choice theory should be universal in application? (pp. 94-95)

One common claim is that “rationality begets ignorance”—”the marginal costs of producing information will generally outweigh the benefits” (which leaves the question of why some people do bother to gather information). In fact, ignorance is widespread even in areas that do not involve collective action problems. (pp. 94-96)

In sum, expanding Olson’s theories to fit the data simply created tautological accounts. The authors offer two suggestions to improve the situation: (1) improved sampling techniques to avoid biased inference, and (2) keeping rational choice accounts analytically distinct from other accounts. (pp. 96-97)

Chapter six looks at rational choice in explaining legislative politics—an essentially thin kind of rationality since the propositions flow from “the geometric arrangement of voters’ preferences” rather that the type of goals they seek. The chapter begins with a theoretical account of instability, cycling, and agenda setting in a committee (the example is an academic committee that sets professors’ salaries). There are two main findings. First, majority rule is “generically unstable” and subject to manipulation. Even if they are not manipulated, majority outcomes are arbitrary, reflecting only the particular circumstances by which one majority overthrows another. Secondly, when the provost appoints someone sympathetic to her views to the committee, the result is cycling that actually makes the provost worse off. (pp. 99-107)

Rational choice therefore produces the hypothesis that legislative preferences are cyclical and subject to agenda manipulation. But this conjecture this is not empirically valid in all cases. Most people would at least expect that instability and manipulation are possible under majority rule—in which case a more interesting research agenda might look at what factors contribute to either instability or stability. (p. 107-113) Alternatively, rational choice theorists look for post-hoc explanations for equilibrium, most notably through models of structural-adjusted equilibrium. Models of committee structure are one example of this, but they are derided by the authors for being “manifestly unrealistic”. (p. 114-120) Laboratory experiments offer possible advantages for researchers because they simplify the complexities of real life situations, and enable research to focus on particular forces at work. But even in the laboratory anomalies arise. Researchers have not gone far enough to probe social-psychological factors that may affect the success of rational choice models. Null hypotheses are too vague, enabling researchers to circumvent inconvenient facts. (pp. 120-123)

Two essays by Thomas Hammond and Garry Miller argued that bicameralism, the executive veto, and the internal organization of the legislature may combine to induce a “core” (an area invulnerable to majority disruption) and hence policy stability. (p. 116) To what extent is the core successful? Experimental research has yielded inconsistent results. Can the core predict successfully? Evidence is inconclusive. The strategic myopia of some players means that the assumption that each individual pursues optimal strategies is unrealistic. Outcomes may be at variance with the predictions of rational choice theory because of the varying game playing talents of individuals. Results are also likely to be influence by the differing ability of individuals to manipulate institutional incentives. (pp. 128-132) Other rational choice theorists have focused on the effects of exogenous institutional changes on actors, with mixed results. Experiments have shown actors to respond in very different ways to the same strategic situation. This heterogeneity challenges the limited axioms about human nature presented by rational choice theory. (pp. 137-139)

Chapter seven discusses spatial theories of electoral competition. Briefly, although most students of American elections would accept that candidates act strategically in elections, far fewer would agree that a Nash equilibrium exists in something so complex and tactically involved as an election competition. Again, the authors call for rational choice theorists to be more open to empirical testing of their arguments. (p. 178)

The book concludes with likely counterarguments from rational choice theorists, and the author’s response to the counterarguments. The counterarguments are listed below:
(1) Naive falsification
(2) No alternate theory
(3) Critics are anti-theoretical
(4) Scientific value of advocacy
(5) Impossibly demanding standards
(6) All theories simplify via abstraction
(7) Rational choice is not one theory
(8) The tyranny of disciplinary divisions
(9) Supremacy of rational choice theory
(10) Critics expect too much of a fledgling theory
(pp. 179-202)

Of course, many of these counterarguments are easily dismissed. And it is certainly true that Green and Shapiro’s book made an impact, possibly in a large part due to the combative tone it takes towards rational choice scholarship. It is also distinguished from other criticisms of rational choice by focusing on the lack of empirical success that the rational paradigm has achieved, rather than the question of whether people can really be assumed to behave in rational ways. However, responses to Green and Shapiro insist that rational choice has enough empirical content to make it , in the words of Gary W. Cox, a “vital and exciting research program”.

In conclusion, the authors reiterate two ways that rational choice theorizing should overcome its problems:
(1) Resist method-based research. Instead of asking “how does rational choice explain X?”, we should ask “what explains X?”
(2) Relinquish the commitment to pure universalism. We should be able to distinguish between rational choice and other modes of behaviour, and create empirical tests to examine what rational choice can actually explain.
(pp. 202-204)

The New England Patriot’s coach Bill Belichick has been taking a lot of flack from Monday morning quarterbacks around the country for his decision to go for it while up by 6 on 4th and 2 from the Pat’s 28-yard line with two minutes left on the clock. Why didn’t they punt? That’s the question everyone is asking because it seemed like the answer was obvious. They should have punted!

Not necessarily. Why? The guys over at Permutations will probably enjoy this: According to computer simulation models, Belichick made the right decision.

The Boston Globe tells the story of Brian Burke:

Burke is a football-crazed, math-inclined single father who works for a military contractor. Three years ago, he found himself with nothing to do once he put his children to bed. “I was pretty bored,’’ Burke said. He used software leftover from grad school to create a simulation tool that could solve football arguments for his friends at work.

…

Late Sunday night, Burke compiled the data and ran the numbers. When Burke began, even though he believes coaches are too conservative, his gut told him Belichick had made the incorrect decision. His outcome proved him wrong.

According to Burke’s tabulation, going for the first down gave the Patriots a 79 percent chance of winning. Punting gave them a 70-percent chance to win. Even after Burke made tweaks, the win probability never dipped in favor of the punt. If anything, factoring in how explosive the Colts’ offense is, the team-specific adjustments only made going for it more favorable.

You can still argue that Belichick made the wrong decision, but at least he had the numbers on his side. More importantly, this shows that mathematical models do in fact have a place in the social sciences. If there are any macro economists reading this, remember that you could always quit and focus your talents on something more productive with far fewer negative externalities for society!

• Josh McCabe

Tsebelis looks at apparently “suboptimal choices.” These choices are puzzling for rational choice theorists who assume actors try to maximize their goal achievement (the utility assumption).
Tsebelis argues that these choices are apparently suboptimal because of disagreement between actor and observer. The observer fails to recognize the optimal action because his perspective is incomplete. The observer is looking at only one game, but the actors is involved in a whole network of nested games (pp. 6-7).
Two major reasons for disagreement between actor and observer:
(1) The actor is involved in games in multiple arenas, but the observer is only focused on the principal arena.
(2) The actor “innovates” to change the rules of the game and create some new available options, so that is better than the original choice. This nested game is referred to as institutional design (pp. 7-8).

Chapter two defends the rational choice approach. There are of course many other approaches to the study of politics. These are grouped into theories without actors (for example systems analysis, structuralism, functionalism, and modernization), or theories with non-rational actors (through impulsive actions or some deeper form of rationality). Translating relationships between different research programs is sometimes possible. If it is possible, but not performed, this refereed to a short-cut or black box explanations. Where translation is not possible, only spurious correlation can result (pp. 19-24).

Tsebelis distinguishes between weak requirements for rationality and strong requirements for rationality.
The weak requirements for rationality are: (1) the impossibility of contradictory beliefs or preferences, (2) the impossibility of intransitive preferences, and (3) conformity to the axioms of probability calculus.
(1) and (2) refer to rational behaviour under certainty, (3) refers to rational behaviour under risk (pp. 24-27).
Strong requirements of rationality establish a correspondence between beliefs or behaviour and the real world. They are: (1) in equilibrium, players conform to the prescriptions of game theory, (2) subjective probabilities approximate objective frequencies, and (3) beliefs approximate reality.
(1) The Nash equilibrium is fundamental to game theory. Players use mutually optimal strategies in equilibrium from which no one has an incentive to deviate. However, there might be more one equilibrium in the game—in this case the problem becomes selecting the best one. If each player chooses a different equilibrium strategy the outcome will not be equilibrium. Deviation from one equilibrium will lead to a series of adjustments leading back to the previous equilibrium or to another Nash equilibrium. (2) Players must alter their probability estimates in line with actual observations. (3) Beliefs and behaviour have to correspond to the real world (at equilibrium) (pp. 27-31).
Deviation from any of the requirements of rationality will result in a reduced level of welfare.

Is there any reason to believe that people actually do act according to rational choice requirements? Milton Friedman offered the “as if” thesis—”it does not matter; people behave ‘as if’ they were rational.” A similar argument was made by Carl Gustav Hempel’s “potential explanation”—the explanation holds if all its premises are true. According to this theory, an explanation may be important even if its premises are not true, because of what it reveals something about the mechanisms that influence the phenomenon under question. Tsebelis argues that the “as if” theory is not satisfactory because it is used to justify “wildly inaccurate” assumptions on the one hand, and produces empirically irrelevant results on the other. He rejects the “potential explanation” theory because the assumptions of theory become its conclusions—from false assumptions “anything follows (pp. 31-32).”

Tsebelis offers a different approach to the question. He does not argue that rationality is a model of human behaviour, but rather a subset of human behaviour. When the actors’ goals and identities are established, and the rules of the game precise and known to the actors, then rational choice explanations will be applicable. As actors goals become fuzzy, or the rules of the game become less precise, rational choice explanations will be less applicable. He argues that “the study of political actors under the assumption of rationality is a legitimate approximation of realistic situations, motives, calculations, and behaviour.” There are five arguments to support rational choice assumptions: (1) salience of issues and information, better approximation to rational choice will be found when the stakes are higher or information is available; (2) learning, people engaging in repeated activities approximate optimal through trial and error; (3) heterogeneity of individuals, social outcomes approximate equilibrium that would be found if all individuals were sophisticated due to “congestion effects”; (4) natural selection, in the long run successful behaviours are reinforced; (5) statistics, even if rational choice is inaccurate concerning a specific individual, it is very accurate concerning the average individuals (this claim is often criticized due to difficulties in aggregating individual preferences).
Rational choice does not explain all behaviour, but only behaviour covered by the five arguments above (pp. 31-39).

What are the advantages of rational choice theory?
(1) theoretical clarity and parsimony, ad hoc theorizing is eliminated, inconsistency between theory and reality is attributed to inadequate theory rather than mistaken actors, strict empirical tests are possible; (2) equilibrium analysis, an important methodological result of the rational choice approach is comparative statics; (3) deductive reasoning, formal arguments (assuming they are correct) are truth preserving, one cannot argue with a theorem (although one can dispute its assumptions), rational choice models can therefore said to be tautological; (4) interchangeability of actors, rational choice does not follow the positivist rejection of Weberian explanatory understanding in the social sciences, but instead argues that “understanding does not depend on any subjective psychological capacity of empathy (this is of course why positivists rejected it) but on the application of strict rules of optimal behaviour under constraints.
The models in the book make two kinds of assumptions: rationality of actors (as defined in the text) and certain institutional structures. Following Lakatos’s “negative heuristic concept,” if explanations or predictions turn out to be false, assumptions regarding descriptions of institutional structures can be modified, but not the core rationality assumption (pp. 39-47).

Chapter three forms the theoretical core of the book. The first section introduces games in multiple arenas, representative of “political context.” Tsebelis shows how payoffs in the principal area are affected by the situation in other arenas or moves players make in those arenas. This can explain the election puzzle introduced in chapter one—the apparently irrational voting patterns that led to the election of Urho Kekkonen as President of Finland in 1956 can be explained by looking at the host of sub-games that parliamentary actors were playing (pp. 53-61, the story of Kekkonen’s election is recounted on pp. 1-5, pp. 53-61).

The second section looks at four simple one-shot two-by-two games: prisoners’ dilemma, assurance, chicken, and deadlock.
T=temptation payoff for player who defects while other cooperates , R=reward for mutual cooperation, P=penalty for mutual defection, S=sucker payoff for player cooperates while the other defects
Prisoners’ dilemma: Ti>Ri>Pi>Si
Each player is better off defecting no matter what the other player does. This leads to a suboptimal outcome as both players receive P rather than R.
Deadlock: Ti>Pi>Ri>Si
Same as prisoners’ dilemma, but defection does not a produce a suboptimal outcome (P is better than R).
Chicken: Ti>Ri>Si>Pi
There is no dominant strategy. Fear of arriving at the worst possible outcome (P) may lead players to cooperate.
Assurance: Ri>Ti>Pi>Si
Mutual cooperation is the preferred outcome.
(pp. 61-63)

In game theory, players will chose mutually optimal strategies. These pairs of strategies are known as (Nash) equilibrium strategies. In the prisoners’ dilemma and deadlock games, both players have the dominant strategy of defection, so the payoffs are P₁ and P₂. The two other games each have two equilibria. In chicken, if player 1 chooses to defect, player 2 is better off cooperating, and if player 2 defects, player 1 is better off cooperating. This leads to two equilibria, T₁, S₂ and S₁, T₂. In the assurance game, if one player chooses to cooperate, the other is better off cooperating, but if one chooses to defect, the other is better off defecting as well. The two equilibria are therefore R₁, R₂ and P₁, P₂. In this game, the R₁, R₂ makes both players better off, so each player can anticipate the other will choose to cooperate. However, in the chicken game, such a selection between equilibria cannot be made—each player prefers a different equilibrium. The divergence of preferences creates a problem of communication, leading to instability. Unless the players can communicate, they may both decide to defect (because they want to force their preferred equilibrium on the other player), or both decide to cooperate (because they were afraid of the other player’s defection).
Another source of instability arises from whether the outcome was Pareto optimal (impossible to improve one player’s payoff without reducing another’s) or not. If an outcome is not Pareto optimal, players know that they can improve their payoffs if they get together—however because communication is prohibited this is not possible. The equilibrium in the prisoners’ dilemma is not Pareto optimal. Tsebelis sums up the argument as follows: “Nash equilibria are outcomes that are stable against unilateral deviations because no player has the incentive to deviate; Pareto optimal points are stable against universal coalitions because it is not possible to deviate from such points without hurting some players (pp. 63-68).”

However, when contingent or correlated strategies are introduced, the properties of the four games changes. In the prisoners’ dilemma, chicken, and assurance games, when correlated strategies (each player uses a different strategy according to his opponents strategy) are possible, the likelihood of cooperation increases with R and S (the payoffs for cooperation), but decreases with T and P (the payoffs for defection). Games in multiple arenas have variable payoffs, and the different payoffs (assuming contingent strategies are permitted) produce the same outcome —regardless of the nature of the game the outcomes depend only on the size of the payoffs (pp. 68-72).

Similarly, Tsebelis shows that in iterated games what matters is not the order but the size of the different payoffs. In the prisoners’ dilemma game a player declares she will cooperate in the first game and continue cooperating as long as the opponent cooperates as well. If, however, the opponent defects, she will defect in all subsequent interactions. If the opponent believes this threat, they will cooperate in each game because they receive Ri rather than Pi in each interaction. The “backwards induction argument” has been used against this discovery. This essentially argues that in a finite game, players will defect in the last round because there is no future to influence. Given this common knowledge, both players will defect in the penultimate round and so on to the first round—so the strategy of both players becomes “all defect.” This can even be shown for games which are finite, even when this fact is unknown. In fact, since humans do not play infinite games, the existence of cooperative equilibria in games with infinite iterations is problematic. However, experiments indicate that players ignore the “backwards induction” prescriptions. The discrepancy between the theory and actual behaviour can be explained by Fudenberg and Maskin’s “folk theorem”—essentially an individually rational outcome (each outcome gives each player no less than she could get by herself) can be supported as a Nash equilibrium of an iterated game. Obviously, when the reward for cooperation increases, smaller time frames are required for cooperation to develop (pp. 72-78).

When contingent strategies are permitted, and infinite equilibria are possible. Rather than being too restrictive a model of actual human behaviour, rationality can produce a greater variety of outcomes than can increase in reality. The core argument is that the chances for cooperation increases with payoffs for cooperation (R and S), and decreases with payoffs for defection (T and P) (pp. 78-79).

Chapter four discusses optimality concerning the interaction between individuals and institutions. The chapter first looks at how people choose optimal behaviour under constraints, and then moves onto why and how people change institutional constraints. Tsebelis argues that if a game varies, it is because of variations in either the payoffs or the rules (or both). Games in multiple arenas focus on the first kind of variation. institutional change on the second. Institutional change can refer to a change in the set of players, a change in permissible moves, and a change in the sequence of play. The term institution refers to “formal rules of a recurring political or social game (pp. 92-96).”

Tsebelis looks at institutions as an investment: “the choice between two institutions can be considered a decision under risk that will generate two different streams of income because the situation is repeated over time.” Like any other investment, a dilemma arises between whether to go for short-term investment with high return, or long-term investment with lower return. Supposing a simple majority is sufficient to change an institution, the majority can then choose between two different options: because institutional change will trigger a series of modifications to other institutions. First, the majority prefer to support institutional arrangements acceptable to the minority in exchange for higher institutional stability (example in Chapter 6 from Belgium). Second, the majority may push through institutional change against the wishes of the minority (chapters 5 and 7 from the UK and France) (pp. 96-100).

Tseblis rejects evolutionary explanations for the emergence of institutions. While it is in the interests of everyone to follow an agreement that coordinates action between individuals (such as which side of the road to drive on), if an agreement promotes the interests of some, but hurts those of others, then individuals will prefer to violate the agreement. Therefore, “under complete information, cooperation between rational, self-interested, independent agents in a prisoners’ dilemma game cannot develop.” Only self-enforcing agreements (pure coordination problems) can survive. Evolutionary accounts are functional in the sense that they lack concious actors. Following Olson (1965), a community of interests cannot alone make people organize and promote these interests (pp. 100-103).

Tsebelis divides between two types of institutions: efficient institutions and redistributive institutions. Institutions are efficient “if they improve … the condition of all (or almost all) individuals or groups in society.” Institutions are redistributive “if they improve the conditions of one group of society at the expense of another.” A graphical representation can be found in figure 4.1 (pp. 104-106).

Efficient institutions Several solutions leading to cooperation in a prisoners’ dilemma are identified. All these solutions violate the assumption of rationality, self-interest, or independence of the payers:
(1) Violations of rationality
(2) Violations of self-interest
(3) Violations of independence: (a) self-reflective and moral arguments, (b) institutional arrangements including facilitating communication and monitoring, permitting binding contracts, modifying the payoff matrix, transformation of the game, creation of an asymmetric framework, and iterations.
These solutions improve the situation for all players by overcoming the prisoners’ dilemma and pushing outcomes towards the Pareto frontier. Institutional arrangements are the more frequent and persuasive explanation for how actors overcome the Prisoners’ dilemma (pp. 106-110).

Redistributive institutions Tsebelis identifies two types of redistributive institutions: consolidating institutions and new deal institutions. Marxists focus on consolidating institutions—institutions benefit the dominant class rather than the whole society. Accounts of new deal institutions focus on the creation of majorities—minorities will bring up salient issues that can disrupt existing majorities (pp. 110-115).

The chapter ends by asking what kinds of institutions are likely to be redistributive or efficient. Tsebelis uses Rawls’ “veil of ignorance” argument to explain the design of efficient institutions. People are uncertain about the position they would hold under future institutions so they “design a society that is better for everyone.” However, because we actually operate with some information about the future, actual institutions are not purely efficient. Perfect information will, on the other hand, produce redistributive institutions—selecting institutions becomes the equivalent of selecting policies or selecting outcomes (pp. 115-118).

Chapters five to seven look at some applications of the nested games model. Chapter five looks at why Labour constituency parties ousted sitting MPs for being sufficiently left-wing, despite the fact that such apparent “political suicide” invariably caused Labour to lose the seat. In a single shot game with perfect information, there is no possibility of conflict between the constituencies and the NEC, or between the NEC and the annual conference. The actor with final voting power simply imposes her will. However, where the game is iterated and information incomplete, it can be rational for left-wing Labour activists to reject their MP—even given the fact that they prefer a moderate Labour MP to a any Conservative. Rejecting a moderate MP can give activists a reputation for being “tough.” The reselection game “is nested inside a competitive game between parties at the constituency and national levels. This competitive game modifies some of the actors’ payoffs and increases of decreases the likelihood of certain strategies.” This case is an example of “motivated irrationality”—what would have been a suicidal strategy in a one-shot game was chosen because the players were involved in an iterated and nested game (pp. 154-156).

Chapter six takes a rational choice approach to consociationalism, taking examples from the Belgian case. The consociational literature generally considers elites as independent of the masses, describes and compromise between different elite groups. If elites compromise, there will be political stability despite societal cleavages (refer back to Katzenstein). However, moves that elites make in the parliamentary arena also have electoral effects. It is therefore more accurate to say that “political elites engage in a parliamentary game that is embedded or nested inside an electoral game (p. 160).”
Moreover, game theory can explain the use of confrontational techniques by political elites. Sometimes, mutual compromise can produce Pareto optimal outcomes, but at other times the Pareto frontier (or in a game of chicken a Nash equilibrium for favourable to one party) is only achieved by alternating compromise with intransigence (pp. 185-186).

Chapter seven looks at the cohesion of French electoral coalitions. France’s two round voting system favours both competition (in the first round) and cooperation (in the second round), making it an interesting case for the nested games approach (p. 190). At the beginning of 1978, opinion polls indicated the Left was leading, but in the election France voted for the Right. Tsebelis argues that the rapid rise of the Socialist party created a first round distribution of votes unfavourable to the Left coalition, generating bad vote transfers and letting the Right in. Mitterrand introduced proportional representation to replace a system that had produced such unfavourable results, but this was reversed after the Right’s victory in 1986. Nested games show (in a way that theories of spatial voting cannot—spatial voting would expect no difference in the quality of vote transfers across constituencies) how cohesion of coalitions increases when it is needed (coalitions have equal vote shares) and politics is visible, but competition between parties increases when all partners within a coalition have roughly equal strength (pp. 231-232).

This post acts as a sequel to last week’s note on rational choice. As part of my Data Analysis class, we were required to read Milton Friedman’s seminal piece, The Methodology of Positive Economics; it is illuminating in certain regards. Specifically, it highlights some issues with the rational choice, positivist approach largely entrenched in economics today.

The additional confounding effects associated with the social sciences- as opposed to the natural sciences- are inadvertently illustrated in Friedman’s piece. Consider, for instance, the example of an entire batch of faulty telescopes, which could be corrected for measurement error (analogous to perception error in the case of human actors), thereby surmounting the problem entailed by the error. However, one could hardly abstract this into the social sciences and assume, analogously, that all human beings think in the same way and can thus be considered as a homogenous amalgam[1]. Yet, this is precisely the approach adopted by Friedman and numerous positive economists when they eschew an investigation of causal processes (Friedman, p. 21). Following the assumption of uniform perception/cognitive processes/errors, Friedman is able to make the logical leap from the individual to the firm: ‘it is only a short step to the economic hypothesis that under a wide range of circumstances individual firms behave as if they were seeking rationally to maximize their expected returns and had full knowledge of the data needed to succeed in this attempt as if, that is, they knew the relevant cost and demand functions’ (Friedman, p. 21). Friedman refutes objections to the tune that assumptions made in economic theory are too far removed from reality; however, he seems to be misapprehending the problem in doing so. The real issue seems to be the universal applicability of assumptions inherent to Friedman’s construct: thus, while the logical leap from a perfect market to a real, imperfect one is not erroneous, the assumption that all human beings would react to the perfect market in logically homogenous ways is[2].

An example might serve to make this point more clearly: consider a Disney cartoon of a princess. While no one would argue that Princess Jasmine accurately represents the average female form, she is a reasonably good approximation within the context of the cartoon. That is to say, when we view Princess Jasmine, we are put in mind of an abstract conception of what the female human being looks like. Friedman’s rebuttal is clearly operational here: while one would grant that Princess Jasmine is quite removed from realistic portrayals of women, she represents an acceptable approximation for the purposes of the cartoon Aladdin. My point is that the real issue is not how accurately the cartoon approximates real human beings, but rather, the acceptability of adopting this approximation as a basis upon which predictions can be made. Although the Arabian sun shines off of Jasmine’s hair, one cannot infer that the same would be true of a real human being under the real Arabian sun. Similarly, while Jasmine’s eyes approximate genuine human eyes, one would not recommend that an optometrist use the cartoon in an anatomical study.

One can abstract from the behavior of a single electron to that of every electron within a system fairly easily; the same cannot be said of perceptive, cognitive human beings, as I noted last week. While these are merely preliminary thoughts on the matter, further inquiry would seek to establish the degree to which theories on individuals could be abstracted into prescriptive theories about larger social entities. Such is the burden of the social scientist and in particular, one seeking to adopt a rational choice perspective.

Do skim through the Friedman piece; his lucid style makes for a quick read.

[2] It should be noted here that Friedman would certainly acknowledge that human actors have differing ends; the issue here is with the causal mechanisms by way of which human beings determine the means to attain those ends, the investigation of which Friedman consciously eschews.

Sociology is sort of like dumpster-diving in that black box: that’s where all the interesting stuff can be found.

That’s Brooke Harrington trying to explain why German drug companies only have directions in German despite the large foreign population in the country. It reminds me of Stinchcombe’s argument of what social scientists should really be doing:

The central trouble with discussions of rationality is that we are taught by economists and decision theorists to treat rationality as an assumption. Only if rationality is an assumption can one derive the mathematical results from which economists and decision theorists make their living. But in the real world rationality is a variable to be explained.

Remember folks, fewer assumptions mean more fun!

• Josh McCabe

This past week, my Perspectives class studied Rational Choice; needless to say, the issue provoked responses from across the spectrum, albeit clustered towards the skeptical end. Indeed, there are problems aplenty with Rational Choice Theory (RC), but I would like to focus only on a specific issue that has been bothering me.

My problem arises from the avowed RC focus on methodological individualism i.e. the individual as the unit of analysis. In this construct, society is merely an aggregate of individual impulses, desires, actions and so on. Thus, from Jeremy Bentham in chapter 1 of An Introduction to the Principles of Morals and Legislation:

The community is a fictitious body, composed of the individual persons who are considered as constituting as it were its members. The interest of the community then is, what?- the sum of the interests of the several members who compose it.

It is important to note here that this RC approach does not preclude formulating a predictive model of society as a whole- indeed, RC advocates repeatedly state their commitment to churning out socially beneficial work. However, implicit within the approach is a belief that societal models can follow from studies of the individual, since all human beings are assumed to display similar principles of behavior viz. an ability to make ends-oriented decisions in the most efficient manner. Therein lies the central assumption about all human behavior that underpins most RC literature: that human beings are purposive, utility-maximizing actors. Notice, this leaves plenty of room open for diversity of ends, secondary motives and so on; however, to the extent that all human beings are purposive, utility-maximizing actors, RC takes a homogeneous view of humanity. Hence, society is an aggregation of individuals…so far so good.

The RC approach is often called the ‘economic approach’ to social science. Now, consider the following from Mancur Olson’s The Logic of Collective Action (p. 161):

Where nonrational or irrational behavior is the basis for a lobby, it would perhaps be better to turn to psychology or social psychology than to economics for a relevant theory.

Is Olson acknowledging that the RC view of human motivations may not always be tenable? It is not a novel concept nowadays, with the rise of political psychology and behavioral economics; indeed, most sensible RC exponents do not adhere to a strictly Bentham-esque view of human society either. The aggregate view of society now seems more a move of expediency (how else would one formulate policy?) than the manifestation of a rigid belief. That having been said, what does a point, such as the one identified by Olson say about the predictive power of RC models? Say the majority of individuals adhere to the RC view of actors, but a sizeable minority does not: think lost causes (Olson’s example), inertia to change, etc. Behavioral revelations on herd behavior and heuristic decision making tell us that the sizeable minority may in fact, be a majority. Regardless, when faced by two (or more) sets of individuals who behave in such contrasting manners, how can one possibly aggregate across individuals? I refer not to statistical aggregation, but the conceptual aggregation that a predictive model seeks to do. This is an important question, given the centrality of predictive power to RC theory.

As far as I can see, RC could attempt to deal with the Olson issue in one of two ways:

1. Seemingly irrational individuals suffer from an information asymmetry that lends their actions the sheen of rationality; thus, in fact, their actions actually conform to the RC theory of society. For example, in the case of lost causes, actors may hold the misapprehension that theirs is, in fact, not a lost cause with a real chance of success.

I find this argument unpersuasive. To be certain, there are instances when a dearth of information can make an irrational actions seem rational (Consider an ant attempting to climb a mountain, the peak of which is not visible to him due to various outcroppings of rock. To a human being (and his larger perspective), the ant is clearly engaging in an exercise in futility…not so for the ant). However, the ant analogy does not extend to every such instance in society- how often have we heard of people “fighting the good fight” because it is the right thing to do. Think of the common trope of the warrior who dies in battle, or the valor attributed to the Light Brigade and so on. Human beings do consciously engage in irrational acts. This leads to the second possible solution:

2. An individual engaged in irrational behavior may have shifted his ends- thus, if I am fighting the good fight, my ultimate end is no longer success, but a desire to go down gloriously. If such is the case, then my continued devotion to a lost cause is perfectly rational.

Again, not very persuasive- does the warrior truly go into a suicidal campaign motivated purely by the desire for a glorious death? More abstractly, can ends always be separated so easily? By shifting the ends, we would seem to merely be engaging in a tautological exercise by way of which, all actions are deemed rational in order to justify their rationality. I am afraid I don’t have a coherent response to this particular “solution”, but it seems intuitively unpersuasive.

In the final reckoning, the crux of the issue is the inability of RC to reconcile its view of human society with instances of conscious irrationality, since it is premised upon either rationality, or unconscious irrationality. My own issue is: even if RC gives a shout out to behavioral explanations of human irrationality, how can its aggregate view of society follow from the same? Any thoughts?

Social Science

STUDIES IN CRITICAL SOCIAL SCIENCES

Haymarket Books is pleased to be working with Brill Academic Publishers to reprint the Studies in Critical Social Sciences book series, edited by David Fasenfest, now available for the first time in paperback. We believe that the U.S. release of the 11 new titles in this series represents an important new addition to Marxist theory and research in the social sciences.

This peer reviewed series includes titles which address topics and themes of sustained interest to critical scholars. These include analyses of global political economy, the environment, and critical research on religion; new research and theory exploring the intersections between exploitation and oppression; and critical engagement among diverse currents of anti-capitalist thought.

The contemporary political moment is one characterized by intense turbulence and crisis, but also of intensifying possibility for those who seek to both interpret capitalist society as well as transform it.

The publication of the Studies in Critical Social Science book series aims to make a contribution to the project of critical scholarship and social transformation.

Culture, Power, and History
Edited by Stephen Pfohl, Aimee Van Wagenen, Patricia Arend, Abigail Brooks, and Denise Leckenby
This volume’s theoretical meditations and empirical studies include essays tracing marketing and advertising to children, the production of gendered subjectivity in maquiladora factories, the racialized economic history of the construction of the Chicago School of sociology, and the normalization of cosmetic plastic surgery in contemporary America.
ISBN: 9781608460434
US $34

Dialectic of Solidarity: Labor, Antisemitism, and the Frankfurt School
Mark P. Worrell
During World War II it appeared that American workers in uniform had all that was required to defend democracy. But did the ideology of the war align with the ideas of American workers? This unique, multidimensional view of the political imagination of the American worker presents a complex study of war and ideology.
ISBN: 9781608460366
US $26

The Future of Religion: Toward a Reconciled Society
Edited by Michael R. Ott
In the socio-historical context of the highly secular, neo-liberal and neo-conservative globalization movement, the question of the social meaning and relevancy of religion has entered directly into the contemporary discourse on the future of humanity.
ISBN: 9781608460380
US $36

Globalization and the Environment
Edited by Andrew Jorgenson and Edward Kick
These essays present environmental dynamics in areas as diverse as Ancient Egypt and the modern Amazon for readers who are new to the world-systems approach and for all those interested in recent efforts to link environmental outcomes and antecedents to global processes.
ISBN: 9781608460427
US $28

Hybrid Identities: Theoretical and Empirical Examinations
Edited by Keri E. Iyall Smith and Patricia Leavy
Combining theoretical and empirical pieces, this book explores the emerging theoretical work seeking to describe hybrid identities— transnational, double consciousness, gender, diaspora, and others— while also illustrating the application of these theories in empirical research.
ISBN: 9781608460359
US $28

Imperialism, Neoliberalism, and Social Struggles in Latin America
Edited by Richard A. Dello Buono and José Bell Lara
This collection of works by prominent Latin Americanists examines the region’s move beyond the neoliberal era. Populism in power, conservative backlash, social movements from below, and the constraints of global capital are all explored in this essential text.
ISBN: 9781608460403
US $28

Liberal Modernity and its Adversaries: Freedom, Liberalism, and Antiliberalism in the 21st Century
Milan Zafirovski
This book seeks to defend liberal democracy from its conservative adversaries and rehabilitates much-maligned liberalism as the ideal system of liberty in relation to anti-liberalism as one of un-freedom.
ISBN: 9781608460373
US $36

Marx, Critical Theory, and Religion: A Critique of Rational Choice
Edited by Warren S. Goldstein
This volume collects essays by scholars employing Marxist frameworks to analyze religion. The authors engage in a critique of positivism, uncritical praise of the market, and one-dimensional conception of rationality of the rational choice theory of religion.
ISBN: 9781608460410
US $36

Marx’s Scientific Dialectics: A Methodological Treatise for a New Century
Paul Paolucci
A thorough investigation into the relationship between scientific method and dialectical thought in Marx’s work, this book brings new insight to the methodological principles that animate his writings.
ISBN: 9781608460397
US $28

Transforming Globalization: Challenges and Opportunities in the Post 9/11 Era
Edited by Bruce Podobnik and Thomas Reifer
This book examines the emergence of a multifaceted resistance to neoliberal globalization. The authors describe how workers, environmentalists, human rights activists, and a wide variety of other groups have united in protest against institutions such as the WTO, the IMF, and the World Bank.
ISBN: 9781608460441
US $28

Race and Ethnicity: Across Time, Space, and Discipline
Rodney D. Coates
In Coates’ collection, scholars dare to contemplate, evaluate and analyze race and ethnicity in radical new ways. They go past the ideological constraints that normally limit such discourse by disciplinarian boundaries and disciplinarian myopia.
ISBN: 9781608460458
US $34

For individual orders, visit http://www.HaymarketBooks.org, or call us at 773-583-7884, or visit your favourite independent bookstore.

Contact Haymarket Books to request exam or desk copies of titles. For exam copies, please enclose payment of $5 for each book under $30 and $10 for books over $30 with your request.

Haymarket Books provides free desk copies to professors assigning a book as required reading. Course adoption orders and trade distribution in the U.S. through Consortium Book Sales and Distribution: phone: 800-283-3572, web: http://www.cbsd.com; and around the world through Publishers Group Worldwide, phone: 212-614-7888, web: http://www.pgw.com

Posted here by Glenn Rikowski

The Flow of Ideas: http://www.flowideas.co.uk

The Ockress: http://www.theockress.com

Disservice (dĭs-sûr’vĭs)

1. A harmful action; an injury
2. An act that is not just

Our health care system is completely devoid of customer service. It is pathetic.

I took my son to have a simple tympanostomy (ear tubes) procedure this morning. I show up, sign in and take my seat amidsts the throngs of people in the surgical center waiting room. I brought my laptop and some reading materials to bunker down for the long wait ahead.

20 minutes later I get called up front to sign some additional paperwork. Instead of being greeted, 15 documents each complete with a full page of legalese is shoved my way regarding various aspects of responsibility, payment, agreement, arbitration, and host of other information. The grumpy lady has clearly done this a thousand times and she has absolutely no tolerance for any of my questions. She paries my first few skillfully, but I don’t let her blunt my questions regarding the finances.

She shows me that the facility is charging me $5,600 but that fee has been reduced by the insurance to $1,799. This is an all in fee for the facility only (includes staff, equipment, monitoring, etc) and does not include fees charged by the physician and the anesthesiologist. I ask what those charges will be (I already knew ahead of time), but she says she is not responsible for their charges and that I would have to speak with those providers about that. I start asking her why they don’t bundle everything into one price so I can compare across various combinations of facilities and providers. She has no idea what I am talking about and ends the conversation by giving me their phone numbers. Take your seat Mister, how dare you ask a question about pricing comes across clearly as she stares me down to my seat.

I immediately pick up the phone and talk to the physician office. After about 10 minutes, I finally get the billing person who is able to provide me the CPT code (69436) and Zip Code (92691) as well as what they charge for procedure ($345). I tell here I am not interested in her price because it is irrelevant and that Blue Cross has already dictated the price that you are going to get. A little defensive, she then relays to me the the administratively set Blue Cross reimbursement that has been dictated to this particular physician ($208.08).  I then ask her about bundling of services and created an Ear Tube product that would include all the components so that I can compare across facilities and providers. She has no idea what I am talking about. I give her the hamburger example (I don’t get separate receipts for tomoatoes, buns, and burger – I get a single price for the thing I want – the complete hamburger). I refer her to Carol.com as an example and she thinks this sounds like a good idea.  When I ask why they don’t do it now that she understands, she says that she doesn’t think the physicians would ever agree to work in that way. She tells me she will pass this along to the physicians, and with a laugh that indicates that will never happen, we end the call.

Next, I call the anesthesiologist group. First the lady attempts to tell me she can’t give the pricing because it is a HIPAA violation. I quickly disabuse her of her ignorance and get her manager on the phone. Anesthesia is unique in all of medicine because anesthesiologist charge for their time in increments called units (typically 15 minutes). So they get a “set up” fee and a “time-based” fee for their services, both in terms of units. So I ask them what their per unit charge is and the manager tells me that it is proprietary information. I call him out on it and say that pricing information is not proprietary, perhaps his costs structure is, but he has a duty to tell me the cost of the service I am about to engage him in. I am pretty frothy at this point and really lay into this guy. He still refuses to tell me his proprietary, negotiated per unit rate with Blue Cross but relents on giving me the overall price. He then passes me along to someone else who looks up in their database and tells me the cost will be either $300 or $360 for the procedure for either a 15 minute or 30 minute anesthesia time. So, knowing they go in 15 minute unit increments, I can tell that there is either 5 or 6 units involved, and therefore a $60 / unit price. So, full pricing is 4 units “setup” and either 1 or 2 units for their time. So much for your proprietary formula and negotiated pricing. $60 bucks every 15 minutes or $240/hour for anesthesiologist time. Thats mid-tier lawyer rates for South Orange County but interesting in how at least this type of physician’s time might be valued by insurance companies.

So finally, after about 45 minutes of phone time, by someone who knows the ins and outs, all the secret handshakes and covert codes, and most aspects of healthcare financing, I am able to arrive at an all in price for a very simple surgical procedures:

Tympanostomy
CPT Code: 69436
Zip Code: 92691
Facility Fee: $1,699.00
Surgeon Fee:  $208.08
Anesethsiologist Fee: $360.00
TOTAL:  $2,267.08

This is great to know the price information for my selected combination of facility and physicians. However, I have no information on outcomes achieved, safety rates, customer satisfaction, or other metrics to determine if I would not be better off with a different combination of facilities and physicians. What do you think the response was when I attempted to ask about health outcomes for my physician?

Pin drop, anyone?

This is not just another rant, but meant to highlight that the very basic, fundamental courtesies expected during a consumer transaction are all but non-existent in health care. Simple things like getting pricing information, like getting helpful customer service, like understanding what you are buying, and the quality features that attract you to purchase something in the first place. Health care should be one area where customer service is impeccable. I believe you begin to see “brands” emerge that get this, invest in it, and deliver it consistently over time. Looking forward to the ongoing retailization of health care – it truly needs it.

Appealing to “culture” to explain a political outcome is a bit like appealing to “upbringing” to explain why a person does something. Unless we are told which aspect of culture is supposed to accoount for the outcome in question, there is no way to begin systematic evaluation of the claim

(Ian Shapiro, Rational Choice, Culture and Problem Oriented Research. PS, Political Science and Politics, 1998).

Tudo que eu precisava saber sobre teoria dos jogos, eu aprendi com RPG.

Pois bem, depois do meu – ouso dizer revolucionário! – ensaio sobre Dr. Spock e o utilitarismo de fundo emotivista nos meio-vulcanos, agora gostaria de finalmente desenvolver este post que tenho anunciado faz tempo – qual seja: Teoria dos Jogos aplicada ou explicada pelo RPG.

Mais uma vez, eu preciso avisar qualquer N00b que tenha desgraçadamente vindo parar neste blog que RPG aqui não é entendido como aquele método de massagem terapêutica exercício para postura (valeu, Marcelo!) ou como uma arma de assalto típica da primeira guerra do iraque. RPG, os chamados Role Play Games, são o passatempo preferido de 5 entre cada 7 dos indivíduos que tu costumava apontar e dizer “um dia desses este maluco vai sacar uma porra de uma faca e matar uns quinze”.

Sigo o baile. Porque RPG?

Por que teoria dos jogos?

Ora, pergunte primeiro pela essência da pergunta do porquê se esconde no abismo da pergunta pela essência, e diante deste abismo te questione pela poiesis desta ilucidez lúcida – mas quase silenciosa no seu fúnebre êxtase – e então encontre no movimento de expansão e contração de uma dialética delirante pela mentira que é o conceito do pensamento fictício que se esconde no Ser do Outro que vem à palavra – quase como uma nascente de prismas belos e opacos numa servidão senil.

Sim?

Nada?

Ótimo, podemos nos mover para o que interessa, que é a matematização do quotidiano.

Fantástico.

Como eu ia dizendo, existe uma série de relações possíveis entre teoria de jogos e RPG, a primeira delas é a noção de jogo-pelo-jogo. Ora, porque alguém entra num jogo? Sei lá. Quem é que ganha um jogo? Que tipo de jogo tem vencedores? O RPG nos coloca numa situação bacana, porque é uma situação onde o jogo é mais importante que os resultados do jogo, e onde jogadores não conseguem jogar para ganhar. Mais ou menos como este absurdo completo que a gente convencionou chamar de “este amontoado de coisa que algum idiota resolveu sacanear e chamar de vida”.

Segundo, RPG é um jogo finito, temos um sistema de regras auto-referente, com coerência interna mais ou menos bem desenhada, e com uma possibilidade de informação total. Como assim? Eu posso saber das possibilidades dos meus adversários e companheiros em um jogo de RPG, mais ou menos como sei destas possibilidades no xadrez. É verdade, temos um número de possibilidades que desafia a memória dos pouco treinados. Mas sabemos, ou podemos saber, todas as possibilidades de um sistema de RPG. Pelo menos se o Ferrari não resolver inventar coisa para acabar com tua felicidade ser criativo.

Neste sentido, primeiro gostaria de explicitar o jogo de RPG como um jogo de RISCO, ou seja, um jogo onde tuas ações implicam em estratégias de jogo pensadas em logo ou curto prazo, para resultados mais ou menos suportáveis do ponto de vista racional.

Por exemplo:

É possível desenhar uma estratégia dominante para uma determinada situação em um jogo de RPG, onde, por exemplo, em um jogo de Exalted um indivíduo Solar agraciado com o signo dos Nights, se encontra numa situação onde está cercado de inimigos. Nesta situação, o Night em questão pode

a) Arriscar um rolamento com 15% de chance de sucesso, e ser totalmente bem-sucedido.

b) Arriscar um rolamento com 85% de chance de sucesso, e escapar com poucos problemas

c) Arriscar um rolamento com 100% de sucesso e tomar no cu, mas sair, vá-lá, vivo.

Podemos então valorar o rolamento a (100), o rolamento b (51) e o rolamento c (1) de acordo com o nivel de satisfação do jogador. No entanto, um jogador racional precisa aqui matematizar seu risco de forma razoável. Em (a), o indivíduo tem 85% de chance de ter uma falha. Em (b), as chances de falha são mínimas, mas o indivíduo não tem a sua satisfação total. Em ( c) o indivíduo simplesmente se conforma com o mínimo resultado possível.

O indivíduo que opta por a, está optando por uma estratégia Max-Max. Ou seja, uma estratégia de Maximização do Máximo sucesso possível (impossível nas outras). Em (b), o indivíduo opta por uma situação MaxMini, onde o mínimo máximo resultado possível é buscado. Em ( c), o indivíduo opta por uma situação Minimax, onde o máximo mínimo pior resultado possível garantido.

Cada uma destas estratégias é defensável, mas seria alguma destas estratégias dominantes?

“Mas Fabrício, que diabos é uma estratégia dominante?”

Nas palavras do professor Clinton, uma estratégia dominante é a escolha de um jogador cujo resultado é sempre bom, pelo menos em um sentido que a situação dele é melhor que antes da jogada, ou diante das outras possibilidades.

Voltando aos Valores esperados pelo jogador temos

a) EV = 100

b) EV = 51

c) EV = 1

Logicamente,

a-b < b – c

-2b < -a-c

b<a/2+c/2

Ou seja, a escolha lógica – dominante – é B, já que ela sempre oferece uma possibilidade maior de satisfação mínima do valor esperado. No entanto, este cálculo presume a estabilidade do jogo, ou seja, que existe uma situação fixa no jogo. Onde os resultados são totalmente previsíveis.

Ok, no RPG os resultados são previsíveis, mas não totalmente previsíveis. Basicamente por causa da interação com outros indivíduos e da chamada sorte.

Tá, então que diabos de modelo dá conta de RPG?

“Dar conta” é um termo forte, mas eu gostaria de brincar um pouco aqui com os chamados jogos sem equilíbrio, que me parece ser o caso do RPG. Porque ? Porque ele é o único modelo capaz de dar conta de um rolamento de dados (sim, isso mesmo!) na escolha da estratégia.

Pensamos dois jogadores. A (x) e B(y)

Cada jogador deve escolher uma estratégia, correto? Escolher uma estratégia aqui significa para A escolher uma LINHA, e para B, uma COLUNA. A esolha de A exclui uma escolha de B, e vice-versa. Então quem joga primeiro é fundamental. Por isso, crianças, é imprescindível comprar pontos de percepção e raciocínio em jogos onde a soma destes define quem tem o primeiro turno.

Neste sentido, o jogo pode ser de pura estratégia, quando a tua escolha independe das escolhas possíveis para B, ou de estratégia composta, quando a tua estratégia é definida pelas possibilidades de B. No jogo que enunciei ali em cima, B vai buscar excluir a estratégia dominante para A no contexto dado, ou seja, a escolha da linha “1”. Para fazê-lo, ele busca ele mesmo a primeira coluna, retirando assim o melhor resultado possível para A . No entanto, isso ainda não expressa o elemento de SORTE inerente ao jogo. Simplesmente o elemento de exclusão de estratégias dominantes. Também aqui, o jogo está CEGO, ou seja, as pessoas não sabem como as outras estão jogando, nem tem certeza do outcome de suas ações.

Na realidade, começo a me questionar se vou conseguir expressar isso de forma adequada através de teoria dos jogos. No entanto, ali em cima eu já estabeleci uma ordem aritmética de prioridades, onde

para A (x,1-x) e B (y,1-y). Agora, vamos adicionar o fator SORTE e ampliar a tabela. Estou mantendo dois jogadores aqui. Porque? Porque estou colocando a coisa não mais nos termos iniciais, de um Night versus um cenário, mas vários jogadores (B) versus um story-teller (A).

Aqui, mais uma vez, o jogo é de exclusão. Para B, temos uma claramente uma estratégia dominante, que seria evitar a coluna da esquerda (EV=-7). A, neste sentido, pode cortar esta coluna das possibilidades de jogo de B, de forma que ficamos com:

Considerando este cenário, B sabe que A deve pegar (se for racional) a linha mais baixa (EV=1). Temos agora:

Aqui, o EV (e podemos simplesmente falar em número esperado de sucessos e falhas, em termos de RPG), não se expressa de fora dominante nem para A ou para B, ou seja, precisamos neutralizar a relaçào entre a e b, para expressar a probabilidade de escolha de linhas por A e colunas por B quando diante deste último cenário.

Minha tabela guia aqui é a seguinte (para A):

Para neutralizar A versus B para A, precisamos equalizar os resultados possíveis, de tal forma que

ax+c(1-x)=bx+d(1-x)

(a-c-b+d).x=d-c

x=d-c/(a-b)+(d-c)

Numericamente:

x = 1-2/1-3+(1-2)

x=1/5

Portanto, para neutralizar B existe 4/5 de possibilidades que A deve escolher a linha de baixo, e 1/5 que ele arrisque a linha de cima. Agora temos, portanto, as possibilidades de x,1-x, quais sejam, (1/5,4/5) e podemos re-calcular o valor do jogo para A de tal forma:

Fazendo a prova real, temos um valor do jogo que é

V= ad-bc/(a-b)+(d-c)

ou

V=-1-6/-4-1=7/5=1(2/5) [q.e.d.]

Agora, vamos largar A e estabelecer o jogo para B. Ou seja, vamos estabelecer o valor de y nesta patota toda.

Vou focar numericamente, apenas para estabelecer a formula, quem quiser verificar os valores, pode ir em frente. É aritmética básica.

Percebam que ay+b(1-y) é o caso, se e apenase se, A escolher a linha do topo, e cy+d(1-y) é o caso, se e apenas se, A escolher a lina de baixo.

Assim, temos em termos de VALOR ESPERADO e para a formulação do valor do jogo para B (se for racional),

ay+b(1-y)=cy+d(1-y)

se substituirmos pelos números acima explicitados, teremos, o valor de jogo de -1(1/5) para B, que é o valor diretamente oposto ao de A . Ou seja, B ganha quando perde! O melhor cenário possível para B é um de perda, mas isso porque A tinha as melhores possibilidades possíveis desde o início, tinha o melhor cenário e as melhores hipóteses de sucesso. Em termos simples: ele tinha as melhores rolagens de dados possíveis, melhores armas. Para B, o jogo é sempre um jogo de administração de crise. Um deus ex-machina.

Por isso, que os indivíduos que constituem B e suas decisões jogam muito mais consigo mesmos do que contra o indivíduo-narrador.

E daí a necessidade de agora movermos do cenário de dilema do prisioneiro. Ou seja, o cenário onde os jogadores estão em situação de tensão e que só podem se beneficiar de colaboração mútua direta – no entanto, existe uma tendência ao absurdo individualista, que acaba obrigando a emergência de um deus benevolente (também conhecidos como Bentley Preto, Fuscão em Fogo, Pássaro Gigante Voador, Capitão Armageddon e Sr. Juizo Final).

Mas agora devo explicar aos colegas no que consiste o dilema do prisioneiro. O dilema do prisioneiro consiste na tua vida. Na minha vida. Na vida de todo mundo. O dilema do prisioneiro foi a forma como Hobbes nos explicou porque invariavelmente tudo que a gente tenta nesta vida dá errado de forma miserável e inevitável. Porque, no fim das contas, não vai dar certo e todo mundo vai nadar em um mar de churrio fedorento, inexoravelmente fadados à destruição mútua.

O dilema do prisioneiro funciona assim:

Entenderam?

Não?

Então deixa eu tentar explicar para vocês:

Vocês são um grupo de quatro jogadores, em uma campanha de Exalted. Indivíduos disciplinados, já criaram um grupo na primeira sessão (o que pode ser explicado por um milagre, tendo em vista que geralmente demoramos pelo menos tres dias de jogo para conseguir estabelecer um grupo, mas estes somos nós). Pois bem, o narigudo narrador diz o seguinte:

Vocês entram na cidade de coloque nome estúpido aqui e entre outras coisas, existe uma movimentação grande de mercadores, vocês conseguem ver um grupo destacado, parece estar carregando uma nobre, talvez coloque nome de personagem obscuro aqui, que tem uma determinada barraca bastante destacada (continua narração)

Um jogador normal, aqui, espera o narrador continuar a explicação, e a quest ser enunciada.

Um jogador normal, claramente, não tem muito o que fazer no nosso grupo.

O que acontece aqui, um dos indivíduos, vamos chama-lo do nome hipotético Filburg O Lançador de Flechas que Esguiam Na Noite Como A Alma Gritante do Sol E Orgulha A Família. Este indivíduo encontra-se entediado pela narração e então diz:

“Eu largo uma flechada de fogo no acampamento da nome de personagem obscuro”

“Tu faz o quê?”

“Eu largo uma flechada de fogo. Vou usar meus charms de archery e de stealth em combo [mostra ficha de personagem] e vou tentar acertar a barraca dela mesma”

“MAS PORQUÊ?”

“Porque eu posso”

“Fabrício, tu vai acabar com o jogo de todo mundo”

“Qual é a natureza do meu personagem, mesmo?”

“Daredevil”

“Pois é, né?”

“Porra.”

“Vou rolar”

* barulhos de dados *

10-8-8-10-8-7-7-4-3-10-8-7-4-6

“Hehe.”

“Ok, gente. O acampamento tá pegando fogo. Inclusive, a porra da vila tá começando a pegar fogo”

Vejam, aqui é uma situação onde o personagem em tela busca o próprio benefício, o seu próprio melhor cenário possível, e ignora solemente as necessidades e prioridades de seus colegas. No caso, se o personagem tivesse agido de forma coletiva, certamente não teria terminado a sessão sendo espancado pelo mesmo grupo que tinha carinhosamente lhe adotado duas sessões anteriores (não sem antes ter sido quase devorado por uma aranha gigante na floresta ao redor da cidade-aquela).

Temos então a chamada situação Best-Worst. O camarada busca o seu melho cenário, mas aaba expondo todos os demais para os seus piores cenários possíveis.

Em um jogo de RPG, esta é a situação clássica. Se indivíduos agirem individualisticamente, as chnces deles levarem todo mundo para o buraco com eles é grande. Ou, deles se beneficiarem individualmente e massacrarem o grupo – o que na long run, dá na mesma que dar um tiro no pé.

Mas então, agir racionalmente em um jogo de RPG é agir de acordo com as necessidades coletivas?
Bem, não necessariamente.

Porque o insight pessoal pode decidir um jogo mais do que o insight coletivo, e já vi situações onde o camarada gritando A GENTE VAI MORRER SE A GENTE TENTAR ENTRAR NESTA PIRAMIDE ESTAVA CERTO, muito embora ele fosse uma voz isolada e chamada de MEDROSA pelos seus companheiros.

Ainda assim, fica claro que a única forma de escapar da situação de dilema do prisioneiro é através de estratégias cooperativas, e quanto maior – e mais diverso -for o grupo, mais dificil fica a situação. Permitem que eu mostre isso com uma DUPLA.

Pois bem, aqui temos quatro situações de jogo. Em 1 temos a posição original de ALGUÉM DEVE JOGAR. Em (5) o cenário é de desastre, em 4, de ganho absoluto de A, 3, ganho absoluto de B, e 2 ganho relativo mútuo. Vamos pensar no seguinte, na posição original, a possibilidade de colaboração é maior.

Inclusive, num jogo de RPG, este é o caso nas primeiras sessões, quando os camaradas querem formar um grupo para permitir que o jogo prossiga. Portanto, as pessoas colaboram. “Eu sou um Dawn, tu é um Night, ela é uma Eclipse, quem sabe a gente forma um grupinho e domina o universo?”

Esta é a parte fácil. Acontece que quando chegamos em (2), os indivíduos não podem mais voltar para a posição original, mas podem com certeza estabelecer um conflito, já que a vai querer se mover para (4) e b para (3). Ambos podem confiar na manutenção da situação de (2), mas com as possibilidades de mudança e ganho, eles podem assumir o comportamento adequado do seu companheiro de grupo?

Percebam, que no meu gráfico ali em cima temos apenas duas pessoas jogando. Imaginem um cenário com 5 indivíduos, onde dois são completamente dementes (vou dar uma pista, um deles costuma nomear todos seus personagens, desde 1995, de Filburg. O outro costuma jogar com uma versão maior e mais agressiva dele mesmo, tendo em vista que ele mesmo já poderia transformar os desavisados em poças amorfas de sangue). Qual é a possibilidade média de cooperação?

Na realidade, tu te surpreenderia. Porque a não-cooperação persistente leva ao não-jogo. No fim das contas, mesmo o narrador que, como demonstrei, está em situação clara de vantagem, costuma colaborar mandando todo tipo de milagre para viabilizar o jogo.

E no fim das contas, este micro-cosmo pode nos dar uma via de acesso interessante para como a gente toma decisões no nosso quotidiano, mas isso é tema para o distropia.

A recent email from the editor of Critical Review, Dr. Jeffery Friedman regarding the (im)pertinence of the rational actor analytic and the spontaneous order framework for sociology, and as I am finishing – for the second time – Professor Peter Berger’s (Dis)Invitation to Sociology, I have put forward these questions for a little web discussion:

1. What is the value of sociology beyond what the humanities and its sister human sciences have to offer?
2. What did you learn in (any) sociology class that you could not have learned in history, economics, political science, etc. class?
3. Is sociology a theoretical and empirical human science or meta-discipline, which is ripe for rational choice trespassers?
4.  Is sociology just a congeries of subject matters or is there a theoretical core that guides the modal number of sociologists in interpreting their research findings?

• Brian A. Pitt

Consegui terminar minha reflexão sobre Rational Choice e Teoria dos Jogos.

Aqui.

Pensando seriamente em transformar isto em um artigo.

Fabrício Pontin

Bom, vou tentar encerrar a minha reflexão sobre escolha racional e democracia com este último post (tu podes ler o primeiro aqui, e o segundo aqui), focando especialmente na questão dos motivos para votar e dos paradoxos que o dilema do prisioneiro trazem para a idéia de democracia.

Pois bem, meu foco agora vai ser no livro de Downs, “Economic Theory of Democracy”, que sugere um problema na pergunta clássica “porque as pessoas não votam?”. O problema em questão é que se focarmos em calculos de benefício direto, votar não traz quaisquer vantagens para o indivíduo. Mais ainda: é mais vantajoso não votar. A pergunta então, tem que ser invertida. “porque as pessoas votam?”, dada a possibilidade ínfima de fazer uma diferença – e a possibilidade de fazer diferença implique em diferenças materiais que a eleição de X ao invés de Y traria.

Partindo de uma série de presunções, Downs vai chegar em uma equação para determinar a racionalidade embutida no ato de votar. Mas quais são estas presunções?

- Voto é uma expressão de valores de rejeição e aprovação de determinados indivíduos. [VpX+{-VpY}]

- Voto implica que o indivíduo se importe com o ato de votar.

- Individuos precisam pensar para além da racionalidade econômica, já que economicamente o custo do voto sempre é maior que as vantagens economicas de votar.

Esta última assunção precisa ser provada matemáticamente. Downs vai fazê-lo através da expressão dos valores do indivíduo quanto à eleição, equacionados com o número de votantes e candidatos. Assim, temos um valor para a probablidade de ser decisivo em uma eleição que é igual ao número de candidatos dividido pelo número de votantes. Este valor p, relacionado com o valor do voto, per se, é então subtraído do custo de ir votar.

Assim,

VpX+[-VpY]=VpV

[VpV-{C}=VeV]

Onde VeV é o Valor economico do voto.

Agora, vamos presumir uma situação normal de eleição, onde os candidatos X e Y são moderados, e os eleitores, na maior parte, reproduzem esta situação de moderação. O valor da rejeição e da aprovação de um candidato sobre o outro vai ser sempre pequeno. Ainda que um indivíduo prefira um candidato ao outro, esta preferência vai ter um valor total expressado em um número pequeno, certamente na casa decimal. Assim, presumimos algo como 99,99n (valor decimal máximo possível) . Colocando cada candidato como moderado, mas na faixa de transição para imoderação (que seria o valor na casa das centenas). No entanto, o custo do voto quase sempre estará na casa das centenas – já que falamos em perder um dia de trabalho, deixar de ir para a praia, gastar com combustivel, este tipo de coisa. Assim, em uma situação normal, o valor econômico de voto é potencialmente negativo.

E, ainda assim, as pessoas votam.

Downs vai dizer que isso ocorre porque o sistema de cálculo ultrapassa o argumento econômico, e inclui fatores de vantagens e desvantagens não expressas apenas materialmente (economicamente) mas também socialmente. Então:

iff (p[VpX+{-VpY}]+[D-{C}])>0, then Vote.

onde

[Vpx+{-VpY}] é o índice de aprovação do voto em um indivíduo X comparado com o indíce de rejeição de um indíviduo Y, chamarei este fator de pV daqui para frente.

D são os incentivos diretos e selectionados (sociais e economicos) de votar (em geral, do ato de voto, não do voto em X ou Y e da rejeição de X ou Y, mas da implicação social de levantar a bunda da cadeira e ir votar)

C são os custos (sociais e economicos) de votar, ou seja, as razões para ficar em casa.

Pois bem, eleitores que votam baseados em ideologia vão focar em pV. Já que para estes eleitores é realmente decisivo que o indivíduo X seja eleito além de Y, se eles não encontrarem motivos para votarem em X , eles simplesmente não irão comparecer às urnas. Isso pode ser decorrente de uma constatação que muito embora o discurso ideológico seja ‘B” ou “C”, a prática é “A”, independentemente do discurso. Pode também ser decorrente de uma estagnação no discurso político que leva o indíviduo a não perceber qualquer valor distintivo na ideologia de X ou Y – isso pode ser bastante ilutrastivo do cenário das eleições americanas, onde poucos eleitores se sentem motivados a irem votar por não verem grande diferença material entre as ideologias e práticas correntes de X ou Y. Este cenário de estagnação é o que chamaremos de cenário típico, e também é um dos cenários mais desejáveis em termos democráticos. Porque? Porque ao contrário do que se pensa, ele não demonstra uma estagnação positiva, mas uma confiança no processo democrático. As pessoas podem votar, mas escolhem não votar por não sentirem uma diferença fundamental. O cenário atípico, aqui, implica em uma polarização radical dos indivíduos X contra Y. No cenário norte-americano, isso parece ter acontecido nas eleições de 2000 e 2004,  no cenário brasileiro, isso claramente aconteceu em 1989, na França, o efeito Le Pen polarizou o eleitorado o suficiente para que individuos votassem em massa em Chirac, especialmente para evitar a eleição de Le Pen, que havia chegado no segundo turno devido a segmentação da esquerda francesa – e também pelo efeito ‘tanto faz’, que possibilitou que uma minoria de eleitores extremistas votassem em bloco, enquanto eleitores centrados votaram de forma segmentada. Na maior parte das eleições, os candidatos tentam jogar com a polarização aritificial do cenário, para motivar mais indivíduos motivados a ideologia a participarem da eleição, criando assim o efeito de “bloco” – política extremamente bem sucedida por Bush em 2004, que manejou o bloco conservador a votar com ele em massa, enquanto Kerry foi incapaz de motivar a base ideológica dos democratas o suficiente.

No entanto, socialmente, verificamos uma pressão para o voto com base nos elementos D e C. Isso porque existe uma parte grande do eleitorado que não é motivada ideológicamente, mas por vantagens diretas  no ato de votar. Os indivíduos moderados não procuram exatamente motivos ideológicos para o voto, mas vantagens em votar. E não tendem a levantar a bunda para votar apenas baseados na sua rejeição de X ou aprovação de Y. Via de regra, este indivíduo vai dizer “tanto faz”.

Daí a necessidade de incentivos materiais para tirar estes indivíduos da cadeira. Sejam eles através de campanhas (Rock the Vote!, por exemplo) ou de criação de mecanismos sociais de isolamento dos não-votantes “a festa da democracia”, “vote ou seja um alienado”. Outra forma de fazer isso é indicar a obrigatoriedade do voto, para obrigar os moderados a votarem – o que seria uma forma de evitar a influencia direta de blocos, mas ao mesmo tempo, tiraria a presunção de motivação pessoal para o voto. Indivíduos votariam, simplesmente, por não gostarem das consequencias de não votar, não pelas vantagens do ato de votar.  Daí a emergência de estratégias como voto de protesto ou anulação de voto, e mecanismos de exploração de eleitores carentes que normalmente não teriam quaisquer motivações econômicas ou estratégicas para votarem – mas que agora são pressionados por interesses locais à comparecerem e votarem.

Com este último cenário temos a situação de um non-zero sum game, onde os indivíduos não participam mais do processo democrático buscando vantagens, mas buscando evitar certas consequências de não participar. Com isso, alcançamos o elemento que Arrow buscava evitar, qual seja: a transformaçao do processo de voto em um processo de jogo. E daí a vedação de Arrow para esta transformação: ela arruina com o andamento institucional da democracia enquanto modelo ideal, e pior: ela está paradoxalmente presente no sistema democrático

Para esclarecer um pouco como isso arruina o andamento institucional, eu preciso voltar para o Arrow por um minuto: Fazemos uma lista de  preferências {4Z.3Y.2X.1A}. Agora, tu SABE que se tu colaborar tu alcanças “y”, caso contrário “x”. Só que aqui, nem X nem Y podem ser decisivos SOZINHOS para o resultado racional, apenas juntos. Por isso eles precisam maximizar suas condutas no sentido de colaboração, e presumir a racionalidade da outra parte.

Vamos transportar isso para um cenário social.

Tu consegues fazer isso se tu tem uma confiança mínima nas pessoas com quem tu divide o convívio social. Nós temos este tipo de situação em convívio social. Claro, tu tambem tem situacoes onde as partes se comportam de forma irracional. A aposta aqui é que quanto mais condições gerais – materiais – de prática democrática tu der para os indivíduos, ou seja, quanto mais o poder público corresponder às concepções individuais de bem que foram subsumidas em uma grande concepção pública, mais o teu comportamento vai poder ser “maximizável”. Por que tu sabe que o outro indivíduo tem uma potencialidade grande de também colaborar. Do outro lado, se tu não sabes das condições materiais deste indivíduo, e de como estas condições materiais implicarem em uma decisão racional [colaborativa], a possibilidade de tu maximizares tua ação é mínima. Porque tu arrisca perder tudo.

Daí a necessidade  de evitar o sistema de dilema do prisioneiro, estabelecida com o fim das estratégias dominantes e inserção de indivíduos que não percebem vantagens materiais no processo de voto. Este indivíduos que normalmente não votam, ao participarem do processo deliberativo acabam tornando este processo um dilema do prisioneiro. Porque para estes não importam as opções conectadas, o voto deles não maximiza para os demais individuos, porque eles votam por medo da não-eleição de X. Em termos gráficos, este indivíduo tem uma hipótese [Best, Worst] e todas as suas outras possibilidades lhe parecem [Worst, Worst]. Portanto, optar por [B,W] é sua única alternativa, senão…

Ao observarmos os processos de voto em democracias não-maduras, verificamos jutasmente uma polarização deste tipo, onde a estratégia dominante do voto de uma parte da população é simplesmente votar em X, e a expressão de alternativas é simplesmente vedada. Com isso, a presunção de colaboração e maximização fica em suspenso até o aumento das condições materiais destes indivíduos que se verificam pressionados para votar, ainda que não tenham motivos materiais para tanto.

Reduzindo em miudos, e em lingua de gente:

O que Downs demonstra, e que Arrow já parecia ter confirmado, é que democracias só são possíveis em cenários onde as condições materiais dos indivíduos são suficientes para eles não serem pressionados ao voto estratégico, ou à intimidação a votar. Claro, estes sistemas de pressão já são melhores que sistemas ditatoriais ou tirânicos onde sequer a hipótese de eleição é dada – mas estas eleições não são, elas mesmas, democráticas.

No entanto, o paradoxo não é tanto do sistema democrático em si , mas da incompatibilidade das proposições democráticas com sociedades extremamente desiguais. Daí identificarmos uma segurança do processo democrático conforme sociedades são mais razoavelmente organizadas.

Arrow e Downs nos demonstram de forma interessante a atualidade do projeto kantiano de maximização do interesse privado, e a leitura liberal – de alguém como Rawls, por exemplo – adquire um contraste mais factual quando orientamos ela por este viés econômico-matemático. Claro, estas assunções são todas construídas, mas ao mesmo tempo, conseguimos verificar o seu funcionamento no dia-a-dia democrático.

Outra coisa importante a ser concluída é que o exercício democrático nada tem a ver com o voto. O voto é um elemento constitutivo de uma democracia, mas não é , ele mesmo sinal de uma maturidade política. Pelo contrário, podemos verificar que a necessidade de votar indica uma perturbação institucional, no andamento institucional, de uma determinada sociedade.

Came across this gem while researching for a faux chi-chi assignment on thought leadership.

Fascinating, all 17 minutes.

Now we know why Idol aspirants ask friends to join them for auditions and Coke rolls out Light, Zero and what-have-you.

In the context of advertising, Giles Rhys Jones says in future, he’ll pair creative concepts with dud options.

Fabrício Pontin

Para quem perdeu a primeira parte, explicando a Social Welfare Function em termos gerais, ela está aqui.

Quando Arrow estabelece a Social Welfare Function, são colocadas uma série de condições para a razoabilidade do padrão de decisão (eleição) desta cláusula. Já coloquei no post anterior um pouco do desenho da SWF e como Arrow vai buscar a razoabilidade do procedimento democrático com base na diminuição do individualismo na escolha do Domínio de alternativas possíveis para serem elegidas.

Pois bem, não tenho maiores problemas com a conclusão de Arrow. É uma conclusão bastante Kantiana: A democracia liberal só é possível, do ponto de vista formal, se indivíduos não pensarem em termos de preferências individuais, mas em termos de preferências maximizáveis.

No entanto, neste segundo comentário sobre Rational Choice e Social Choice (ainda tenho mais um engatilhado), quero focar na condição de independência de Arrow. Meu ponto é que esta condição demonstra um déficit na compreensão temporal (linear) ali expressada, e vou tentar argumentar que se adotamos uma compreensão mais dinâmica de tempo podemos modalizar esta condição de tal forma que ela perde o sentido.

A condição é assim expressada por Arrow:

Condition 3: Let R_1,…,R_n and R’_1.,…, R’_n be two sets of individual orderings and let C(S) and C’(S) be the corresponding social choice functions. If, for all individuals I and x and y in a given environment S, xR_iy if and only if xR’_y , then C(S) and C’(S) are the same (independence of irrelevant alternatives).

Em lingua de gente, estamos tratando com o seguinte cenário:

Se observamos o cenário R e o cenário R’, verificamos que o candidato “x” vence a eleição ainda que o candidato “y” desapareça no cenário R’. Para Arrow, este cenário demonstra a irrelevância dos candidatos menos preferidos em um jogo democrático. Já que se retiramos um candidato que não tenha sido mais preferido, o resultado não muda. Na concepção de Arrow, esta condição é fundamental para que se evite o jogo estratégico e a formação de alianças que possam prejudicar a eleição do candidato factualmente preferido.

Arrow contra-coloca à este modelo o modelo de “Board-Voting” ou voto elencado. Onde tu podes elencar o teu grau de preferência por este ou aquele candidato. Com o modelo Board, Arrow vai acusar um paradoxo na eleição, por que a condição é violada. O que acontece:

Em R, é eleito X, com grau de preferência 10. De forma isolada. Em R’, temos um empate entre X e Z, com 7 pontos. Arrow vai apontar este efeito como profundamente anti-democrático, já que o candidato preferido torna-se “menos” aceito com o desaparecimento de uma alternativa irrelevante.

Além disso, surge com o board voting o problema da votação paralela, e da formação de alianças. Indivíduos podem votar em seus candidatos “menos” preferidos para não permitir a eleição do seu candidato “mais” rejeitado. Por exemplo, no caso acima, o indivíduo R2 poderia ser decisivo para a eleição simplesmente mudando o seu voto de XpYpZpW para XpYpWpZ, ou,em R’2,  XpWpZ oposto a XpZpW.

Arrow vai indicar que esta possibilidade vai bagunçar todo o processo democrático, já que faz indivíduos abrirem mão de suas opções “sinceras” por opções “estratégicas” e transforma o processo de eleição em um jogo aritmético. Jogo este que é possível já na hipótese de  R, mas é mais flagrante ainda em R’.

O problema desta tese é pensar que temos em R e R’ a mesma eleição. No momento que o candidato “Y” é retirado das possibilidades de voto, qualquer indivíduo racional terá que repensar sua ordem de preferências e atribuição de prioridades. Ainda que ele equalize esta ordem de forma similar à do cenário anterior, não estamos, de forma alguma, lidando com o mesmo cenário.  A ausência de “Y” é historicizada pelos indivíduos votando, e quaisquer que sejam os contextos do desaparecimento deste candidato, ele também participa do processo de decisão. Não é necessário que a contingência “desaparecimento” implique em uma irrelevância, pelo contrário, parece que a Arrow mostra mesmo o quanto o desaparecimento não é irrelevante. Ele causa uma mudança fundamental no processo de escolha, e não compartilho da preocupação de Arrow pela mudança nos resultados.

Também creio que a questão da votação “sincera” oposta à votação “estratégica” merece alguma reflexão. No que consiste uma votação sincera? Até que ponto indivíduos que estão fazendo opções podem ser realmente sinceros com relação ao que sabem das propostas de candidatos x, y,…,z? Arrow mesmo demonstra como um padrão de voto sofisticado pode mudar esta questão de voto estratégico, afinal, uma opção baseada em rejeição pode ser tão sincera quanto uma baseada em aprovação :

::eu desgosto mais o candidato “y” do que apoio o candidato “x”, então é perfeitamente razoável que eu vote no candidato “z”, que eu nem gosto tanto assim, para que o candidato mais rejeitado não seja eleito. ::

Como esta hipótese não é racional? Mais ainda, como posso dizer que ela não é sincera? Parece que Black tem razão ao sugerir que devemos largar a hipótese do voto sincero e focar no voto útil. Mais que isso, eu sugeriria que devemos abandonar também a concepção linear de processo eletivo que Arrow segue. Indivíduos historicizam suas perspectivas ao fazer opções de voto – para o bem, ou para o mal. Esta historicização pode ser desastrada, pode ser completamente irracional, mas ela, sem embargo, ocorre.

No próximo post vou voltar para esta questão do voto útil, para concluir esta minha primeira reflexão em Rational Choice. Mas daí vou migrar do livro do Arrow, para o trabalho de Downs.

Fabrício Pontin

Foi Gödel que demonstrou como alguns sistemas, a partir de suas próprias premissas, provocam a indecidibilidade do seu propósito geral. Para Gödel, o problema estava em Russerl, que havia deixado pontos incompletos ou inconsistentes na sua teoria matemática. Gödel foi capaz de demonstrar como organizações aritméticas podiam causar paradoxos, no sentido que um sistema que procurava provar P acabava por levar a ~P.

Arrow, no seu Social Choice and Individual Values parece apontar para uma indecidibilidade das proposições vinculadas à democracia. Como isso ocorre? Isso ocorre porquê um modelo ideal de democracia busca eleger um candidato representativo dos valores dos indivíduos que votam de uma forma que possa ser identificada como não-impositiva. O primeiro problema que encontramos, então é o chamado Paradoxo do Voto.

{fav} é o candidato favorito

{med} é o candidato secundário

{rej} é o candidato rejeitado.

Digamos que indivíduos (1), (2) e (3) expressam seus valores e gostos individuais de forma adequada ao votarem nestes indivíduos na hierarquia que eles estão apresentados, e vamos presumir que estes indivíduos estão expressando esta hierarquia de forma consistente, de forma que não exista circularidade na hierarquização.

Assim, o indivíduo (1) prefere A ao B, B ao C, e se for um indivíduo racional irá necessariamente preferir A ao C, pelo menos se assumirmos que ele expressou sua hierarquia de forma sincera no seu voto. Pois bem, se presumimos o mesmo com indivíduo (2) e (3), o paradoxo apresentado é que em uma eleição onde todos votaram de forma racional, e de acordo com seus interesses, ninguém foi eleito.

Ninguém foi eleito pois todos os candidatos são igualmente tidos como {fav}, {med} e {rej}. Assim, a única forma de resolver a eleição é um critério de desempate, que também precisa ser votado – e é ele mesmo exposto ao mesmo paradoxo do voto acima.

Ou seja, a democracia carrega nas suas premissas a possibilidade de sua negação lógica. Existe um cenário possível, onde nenhuma decisão é tomada pelo voto direto.

A pergunta de Arrow, então, é se existe um procedimento democrático possível onde este tipo de paradoxo não é existente, onde a democracia se torna uma proposição decidível.

Em primeiro lugar, é preciso notar que Arrow quer evitar as condutas do tipo instrumental neste contexto. Se seguimos o autor, não há como compatibilizar a idéia de jogo com a idéia de democracia. As partes que estão votando devem representar seus interesses de forma não-estratégica, mas sincera. Por sincera, Arrow entende uma conduta onde as preferencias individuais são mantidas de forma autônoma e são realmente representadas pelo candidato no qual o indivíduo vota.

Assim, temos:

(x,y,z)- candidatos

S= set de todos  as alternativas possíveis

P = Preferência

I = Indiferença

R – Preferência ou Indiferença

c(s) – ‘Choice Set”, ou alternativas escolhidas.

Deste modo,  temosas seguintes definições

(D1)

xPy <-> ~yRx

(D2)

xIy <-> {xRy + yRx}

(D3)

xRy <-> ~yPx

Até agora estamos lidando com definições de posições individuais, que ainda não estão em relação de preferencia com outros indivíduos. Não existe, ainda uma mediação institucional ou inter-subjetiva das preferências para uma decisão “democrática’, mas apenas a formação de um julgamento de valor (se estamos lidando com um Liberal) ou de um gosto (se estamos lidando com um Utilitarista)  que é expressado de forma Consistente (D1), Coerente (D2) e Direta (D3). Pois bem, agora vamos complexificar nosso problema trazendo o ordenamento de relações

A função (f) expressiva do bem-estar social é aquela que é igual ao ordenamento de valores (Rx) de um número de individuos (x) onde [Rx=f{R1, R2...Rn}]. Assim, “Rx” satisfaz os valores/gostos de um número de indivíduos participantes no processo de decisão social de forma direta, e presumindo que os indivíduos expressaram seus valores de acordo com (D1), (D2) e (D3) a Função do Bem-Estar Social poderá ser também chamada de razoável.

O problema é que o set de alternativas possíveis (A) é enorme, e o conhecimento sobre todas preferências e valores possíveis para os indivíduos no processo de escolha é impossível. Ainda assim, podemos presumir que em um set de várias escolhas possíveis (A) temos um set menor de escolhas disponíveis (S) no momento de votação. Como não podemos descartar sets de preferencias ou estados sociais aprioristicamente no processo de escolha social, temos uma cláusula de domínio (S) que quanto mais abrangente mais causa paradoxos (já que ela tende ao infinito, se todos os indivíduos pensarem de acordo apenas com suas preferências pessoais).

O outro problema que a cláusula de domínio traz é que quanto mais indivíduos ampliarem o escopo de seu numero de preferências, maior é a possibilidade de um indivíduo (i) ou grupo (g) determinarem  (D) a escolha social [C(S)].

Digamos que temos um número de indivíduos decidindo (N), e um grupo pequeno (g) e outro grupo majoritário (N-g). Presumindo a seguinte ordem de preferências:

I)

(a) xPy

(b) yPz (por vedação à imposição de N-g em ‘g’)

(c) xPz (por transitividade)

(d) gDxz (Q.E.D.)

II)

(a) zPx

(b) xPy (non-I)

(c) zPy (Trans)

(d) gDzy (Q.E.D.)

Então,

[iff] gDxy <-> gDxz <-> gDzy

Assim,

gD[g(N-g)]

Finalmente,

gD[C(S)]

Isso acontece porque o número de possibilidades de domínio é tão grande, que um grupo pequeno pode ter um set de prioridades tal que é colocado de forma sobreposta aos demais votantes.Se assumimos algum cenário onde gD[g(n-g)], e precisamos presumir esta possibilidade em uma sociedade democrática, então segue, conforme demonstrado acima, que um grupo sempre pode determinar  o resultado final, e quanto maior for o domínio de candidatos e preferências possíveis, maior a possibilidade de difusibilidade, ou seja, de uma representação menor ser a eleita. Arrow chega a demonstrar que mesmo um indivíduo pode em um cenário hipotético decidir uma eleição democrática de acordo com seus próprios valores individuais.

Arrow demonstra bem como a liberdade de escolha de valores e prioridades precisa ser mediada com um sentimento de igualitarismo e solidariedade para que as estruturas democráticas possam funcionar de forma adequada. O interessante é que Arrow é capaz de nos sugerir, ao final de sua brilhante tese, que apenas se os indivíduos [N] expandirem seus interesses indivíduais para que estes possam ao mesmo tempo serem interesses públicos a democracia é possível.

O interessante do Arrow acaba sendo o quanto esta tese é informativa para quem deseja compreender o que está em jogo na idéia de véu-da-ignorância e posição original em Rawls. Parece que Arrow foi profundamente influente na tese de Rawls para a construção de um liberalismo político.

Claro, Arrow não foi capaz de desenvolver de forma satisfatória os elementos epistemológicos com os quais ele está lidando ao falar de “valores”, “expressão sincera de valores”, “expressão estratégica de prioridades” ou mesmo “irrelevância”. Ainda assim, é difícil não concordar com alguns dos pontos demonstrados pelo Economista americano, que ganhou um nobel por este livro.

Da minha parte, minhas principais resistências são ao que Arrow chama de “cláusula de independência” e à um certo déficit idealista na descrição dos processos de escolha. Existe uma certa linearidade na compreensão de expressão de valores por parte de Arrow que me incomoda. Quero dizer, se retiramos um candidato de uma eleição e expressamos novamente nossa hierarquia de voto, estamos diante da mesma eleição? Parece que o problema aqui é de uma compreensão linear na forma como escolhemos candidatos – diga-se de passagem, esta compreensão linear de tempo e expressividade é uma tendência da filosofia analítica que indica bem o déficit do qual falei ali em cima.

Os colegas podem ler o livro do Arrow aqui. O link é da Yale, então não é pirataria. Podem ficar tranquilos.

Ah sim, boa parte das minhas derivações lógicas são baseadas em anotações que fiz na aula do professor Robert Clinton. Portanto, crédito para ele nas modalizações que eu tentei reproduzir aqui nas minhas limitações matemáticas.

Everybody who has tried to overcome habits like nail biting or checking email twice an hour knows how difficult it can be to break a habit.

Vice versa, if one hopes to introduce a new behavior into ones life, for example to work out regularly, making it a habit is a powerful and effective way to do so.

The best part: if you know a little bit about habits, it is not very hard to develop, change or drop habits. There are hundreds of blogs, websites, books and personal trainers who specialize on that field. ZenHabits is one place to go, this is another.

This post was triggered by me stumbling upon the following display of (potentially) habitual behavior while reading the International Journal of Game Theory (working on my thesis):

In the two player public good game, each player receives some money and then decides on a part of this to be invested into the public good. For each dollar he invests, he receives 0.75 dollars back and the other player receives the same amount. Investment into the public good thus increases the total payoff, but not the individual payoff of the investor. While superrationality or altruism explain cooperation in this game, experiments have shown that many players invest substantially even when they know that they are playing with a phantom partner and that their decision does not effect any payoff but their own.

A possible explanation is that individuals decide based on learned habits, rather than the situation at hand. Since cooperative behavior is usually rewarded, many humans learn to cooperate and continue to do so regardless of the game played.

When asked, most people would prefer a payment of 1 billion US$ to a lottery that gives them either 2 billion US$ or nothing, both with a chance of 50%.

This is rational choice, if you appreciate that the difference in utility between having nothing and having a billion dollars is actually quite large, while it does not matter as much, whether one person owns one or two billion dollars. They are vastly rich in both cases.

The personal utility lost when giving away the first billion outweighs the personal utility gained by the transition from US$ 1 bill. to  US$ 2 bill.

While the expected value of the lottery equals 1 billion dollars, the expected utility is lower than the utility of the risk-free payment. This behavior is known as risk aversion.

It follows that a person who bears financial risk – depending on his personal risk aversion – can increase his utility by buying insurance against the risk, even if the insurance company charges considerably more than the expected cost and makes good profit.

My next work on behavioral economics is about  pricing effects in changed economic conditions. Here is a sketch of the ideas I am working on. Here I introduce a new phrase that mirrors “Willingness to Pay” – Wherewithal to Pay. While you are there do check out my other decks as well.

Twitterview (twĭt’ər vyū) n.

1. A twitterview is a combination of the terms Twitter and interview.
   
2. The Twitter medium of 140 characters forces a concise style of interviewing and response.
   
3. The public can join in on the conversation and become participants themselves by following along or tracking hashtags.

On March 26, 2009 the leading health care bloggers (see list below) throughout the blogosphere participate din a Blog Rally to raise awareness for public participation in the Healthcare X PRIZE design. Bertlan Mesko, leading Medicine 2.0 Advocate and author of the popular Science Roll blog, also conducted a “Twitterview” in support of the effort.

Berci: Can we start the twitterview now? I’d have 10 short questions, you may have 10 short answers. So everyone can enjoy it.

HealthXPRIZE: Thanks for taking the time. We appreciate your help in getting the word out. This Twitterview will complement the Blog Rally. Ready!

Berci: Great! First, what is the X PRIZE Foundation? What is the X PRIZE model?

HealthXPRIZE: The X PRIZE Foundation is a non profit organization that conceives and operates large incentivized prizes that lead to revolutionary breakthroughs. The X PRIZE model is based on leveraging a large purse, with a clear set of rules, that allows innovators to break through barriers.

Berci: Please tell us more about Healthcare X PRIZE!

HealthXPRIZE: The Healthcare X PRIZE is intended to be a competition to redefine health and demonstrate how new models of care can dramatically increase health value. We chose to focus on health value as opposed to a new wonder drug or device as our sponsor (WellPoint and WellPoint Foundation) & advisors were most interested in a systems prize. Systems prizes are much more difficult to conceive and operationalize than technical competitions like going to space or even replicating the genome rapidly. We are expecting that teams will need to innovate around health finance, care delivery, and individual incentives to increase health value. We are currently developing a clear set of rules, which provide the parameters of competition, as we believe that “creativity loves constraints”.

Berci: Reforming the US healthcare system is quite a brave mission, isn’t it? Why the focus on health value?

HealthXPRIZE: The US Health reform gets serious this summer and the HXP is well timed to actually demonstrate and prove in practice the principles of reform. Value is powerful organizing principle for reform efforts – we cannot just reduce costs, nor can we just attempt to improve quality without financial accountability. The focus on health value highlights the need to focus on both sides of the equation. Since Value =outcomes/cost, we are challenging teams to improve both simultaneously.

Berci: Why use an incentivized competition?

HealthXPRIZE: Incentivized competitions are very efficient, highly leveraged, and create an “X” factor within the competitive framework. Sponsors only pay the winner, a $10MM purse typical spurs >$100MM of investment, and the X factor creates global media attention to a key problem, inspire hero’s, encourage non-traditional thinking, and creates a powerful incentive for innovation.

Berci: And how can you properly measure health value? I guess you need pre-defined parameters. What are these?

HealthXPRIZE: Health Value has never really been measured within the US Health Care system. There are many efforts underway right now to properly define and measure health value. Many innovators are leading the way and we are attempt to build on their work or actively collaborate with new/ongoing initiatives (Dartmouth, IHI, AHRQ, etc) to solidify the health value measurement framework. In the context of competition, we are trying to make our measurement framework as concrete as possible by focusing on outcomes (mortality, specific morbidity, ED visits, hospitalizations, sick days etc.). Effectively communicating the notion of “health value” remains a challenge; we are considering focusing on aspects of health value (like decreased hospitalizations and sick days) as a more effective way to communicate to the public the hoped for prize breakthroughs.

Berci: How are the Teams and Test Communities Selected?

HealthXPRIZE: Teams will be selected by through a series of concept design and testing evaluations. They will be required to demonstrate or model the impact of their proposed interventions against test database provided by WellPoint. Independent judges will evaluate the merit/validity of the concept in order to advance. Communities will be selected based on specific criteria that are still being worked through. Intent is to have a defined population of 10K participants from which Teams will voluntarily enroll in the intervention. Test community will be matched against a geographically adjacent control group. Both the team and community selection requires further design, detailed analysis, and expert opinion which we are soliciting at this time through our network of national measurement experts.

Berci: When does this competition start and when will it end?

HealthXPRIZE: The “competition” has several phases: Design, Selection, Competition. We are currently in Design phase through our anticipated Launch later this fall. The Design phase includes soliciting public comment on how we can improve our initial concept/construct to create the most viable competition possible. After official “Launch”, we will begin recruiting teams to compete. Teams will then be narrowed as described above through late Spring 2011 when 5 finalist selected. After a brief integration period into test community, HXP competition is planned to officially begin in January 2012.

Berci: How does this shift the paradigm? What kind of outcome do you expect?

HealthXPRIZE: Great question – we believe the current paradigm is based on volume not value, on process not results, and incents the wrong behaviors while delivering bad outcomes. We want to shift the paradigm to rewarding the reduction of hospitalization / sick days and begin to pay for overall health improvement (this is the outcome we want!). We also want to not focus solely on disease care, and aren’t interested in just improving health care; but believe that we must move to an entirely new notion of engaged, activated health called “Vitality”. We want to demonstrate that this CAN be done at scale, with new entrants / new ideas, and want to set the HXP up as a framework from which these efforts can be tackled in the real world. By focusing on outcomes, instead of regimenting care processes or dictating care delivery, let providers/patients innovate and create rewards for those who obtain the best outcomes.

We believe incentivized competitions are a great vehicle from which we can accelerate change, shift the paradigm, and be a catalyst for the transformation that is required for the US healthcare system. We hope the outcome is a new way to think about health, measure health value, and demonstration of new models of care that demonstrate how to improve community health and individual vitality.

Berci: My last question, regarding X-PRIZE – first rockets, then genomics, now healthcare. What do you think? What’s next?

HealthXPRIZE: XPRIZE is a mission driven organization seeking to inspire the very best in human kind for the benefit of all – this isn’t just a nice quote. It is inherent in the DNA of the organization. We are attempting to be the catalyst in any “stuck” industry by creating incentivized competitions that can lead to radical breakthroughs to the grand challenges of humanity. HXP is now looking at education, energy (some really cool stuff), and developing world initiatives that can truly have major impacts. Fortunately for me, HXP is our focus for launch this year. It is quite challenging work, deals with multiple hard to think through issues, but includes the privilege to work with great people and teams including our sponsor WellPoint.

I have been thrilled with the level of commitment to this process and this prize development process has been tremendous experience. They have a very talented innovation team, led by Chad Pomeroy, who is fully supported by senior executives all the way up to Chief Executive Officer Angela Braly. They have been driving this initiative forward far beyond the $10MM prize purse; they are providing operational resources, sharing data, working to create appropriate test communities, altering business practices to accommodate the prize, and are committed to transparency as part of the HXP process. Their commitment to the project is the reason I became involved as I saw an unprecedented opportunity to really implement the innovation in an idealized but competitive test environment. We appreciate WellPoints leadership, foresight,and commitment to engage X PRIZE in developing the Healthcare X Prize for benefit of all. Very cool stuff.

Berci: Thank you very much for the interesting answers! I will publish the transcript on Scienceroll.com in a few minutes.

HealthXPRIZE: Berci, again, thank you for this twitterivew. We hope to have everyone visit our website, download the initial prize design, comment on our blog, and add their input to the Prize Design process.