Although the late modernity, the present one focuses on individuality and maximum freedom, game theory, although built on the same principles, can provide valid propositions only in well delimitated cases. Here we are in front of a metatheory problem – understanding the fact that although the pre-requisites of a theory are consonant with the pre-requisites of the social fact, the theory has a reduced validity degree in what concerns understanding the way social order is generated. This fact owed to the essential changes generated in structure of other components of late modernity, as delocalization of space and the time induced pressure.
The Problem:
How are social norms generated? Why do people choose particular types of behavior and how do some of them become social norms?
Functionalism, but also game theory gave explanations of this aspect. In the first case, the individual behavior is explained throughout a systemic approach, in the second, we talk about the individual and his dyadic relationships. The lacks of the functionalist theory in explaining the way in which social norms are generated are already known. They refer to the fact that the individual is seen as Homo sociologicus, and his behavior serves the group / the organization he belongs to. In the case of game theory, we deal with a Homo economicus, oriented towards maximizing utility.
The Question:
Can we accept as valid the pre-requisites of game theory and, implicitly, of the rational choice theory in the context of the present modernity? In other words, are they extensible to the level of the entire society? Are they a major component of this modernity?
The assumption of this analysis is that giving the changes in what concerns the rationality in the present modernity, the explanation area provided by the game theory reduces. The concepts currently used in game theory, such as CKR (common knowledge of rationality) and CAB (consistent alignment of beliefs) are harder and harder to identify and the framework of systemic rationality seems significantly reduced. This doesn’t lead even by far to repudiating the theory in favor of a functionalist view. Knowledge is not the expression of a zero – sum- game. We only want to emphasis the explanatory limits of the given theory. For this, we will analyze 3 static games (prisoner’s dilemma, rabbit or stag, hawk and dove) and a dynamic one, illustrating the problems each game reveals. We will compare these problems with the aspects emerged from the speech analysis on the present modernity.
Reasoning:
CKR represents one of the main pillars of game theory. It assumes that, giving the fact that A is more important than B, and B is more important than C, every logic conducts us to consider that A is more important than C. CAB completes CKR. If a subject uses CKR, we can assume that the person with whom he interacts will follow the same CKR logic, thus the first subject is able to foretell the way in which the second defines his decision options and thus, his behavior. Maximizing the utility, each player will choose those types of behavior that will bring him bigger gains. In which way the players will cooperate, so that each one to be satisfied, transforming their choice and their behavior into social norm?
We will start our analysis with the famous “prisoner’s dilemma”, a static game, defined by the Nash equilibrium. The game states that all players gain if each player receives something and doesn’t give anything in return (the free-rider effect), that each player will lose if he offers something without receiving anything in exchange, and if both players want to receive, without giving anything in return, they will both find themselves in the worst situation.
Table 1
A°
| |||
competition
|
cooperation
| ||
*B
|
competition
|
*5, 5,°
|
*0, 10
|
cooperation
|
10, 0°
|
2, 2
| |
The numbers in cells indicate years in prison. The best solution for each player is to defect from their implied pact and try to take advantage of the other one’s loyalty, in which case the one that betrays gets read of the sentence (0), and the one that keeps his word gets 10 years in prison. This free-rider behavior is possible only if inside the society there is some kind of morality related pressure, similar to the one described by Weber in “The Protestant Ethic and the Spirit of Capitalism”: “the good paymaster is lord of another man’s purse”. In other words, while the moral level requires cooperation, at the pragmatic level wins the one that betrays.
But if the immoral behavior generalizes (in other words, if only the pragmatic aspect becomes important) and everyone brakes the promise in order to maximize their interest, everyone will suffer. How can one avoid this? Either morality is very strong, or significant costs related to the abandonment of loyalty and entering in competition appear, which compel us to maintain cooperation. In weberian terms, we find ourselves either in a situation defined by value-oriented rational actions, or in one characterized by goal oriented rational actions. In the first case, concepts such as culture, mentality, anterior learning, socialization, are not related to the game theory bases, which focuses on utility, thus we ought to abandon this direction. This means that in those particular cases that fallow this logic (an example could be the modern rationalist fundamentalism), game theory cannot be used.
Let’s analyze the problem in terms of a goal oriented rational action that has, at least hypothetically, the same ground as game theory. What would determine the players to respect their commitments in terms of CKR and how could this become a social norm? The answer could consist in the direct sanctioning of betrayal (analyzed by Axelroad and defined as the Tit for Tat strategy) or by sanctioning from a third party. The reciprocal sanctioning of betrayal depends on what each player believes about the other one. The equilibrium (solution) will depend on the information the players have about each other and about their behavior in a specific situation. We can assume that this information is provided by the group players activate in directly, using stories, metaphors, or indirectly, through rumors.
In real terms, are these requirements respected by the players? Let’s take an example: we go to the market to buy some cheese, behavior defined by the notion of “utility”. Besides the buying cost itself, we also have a transactional cost, related to the process of collecting information about the quality of the cheese. Let’s assume that we identified a good cheese and the first thing would be to find out who is the producer, so that we can buy from him another time too. The experience tells us that once we took this decision, in a similar situation we should buy from the same producer and we should repeat the behavior until the choice will prove itself fatal. In that moment we will re-engage in the practice of evaluating the other producers, minus the first, about which we already have the information that he is not a safe one. If the society promoted the idea according to which all cheese producers are interested in offering a superior quality cheese and if they did everything possible in this respect, then , even in the case of failure, trust wouldn’t be broken and the client would go to the same cheese producer even the third time. We can assert that the cultural orientation towards quality would allow some savings in what concerns the transactional costs, even if some accidents may appear. Therefore, the quality oriented producers would be “saved”, even if they failed. Usually, their strategies are related to “brand” building that “hides” a “safe” quality and which allows making transactional savings. Therefore, inside a society with high trust capital, people would not take pure CKR choices.
Question: in a low trust society, should people’s choices base on CKR? In this case, the quality oriented producer, even if he fails once, doesn’t benefit from resources from the client, the way it happens in the first case, and his ulterior efforts towards obtaining quality will not be rewarded because the doubting clients decided to start spending again resources to obtain information and because of this, they don’t believe that there could be a quality oriented producer. Moreover, there are enough producers and, implicitly, a significant quantity of liberty. Therefore, this producer, in front of those additional costs, is tempted to give up the quality improvement related costs and to produce goods without pursuing quality, amplifying in this way the client’s confusion. The entire demonstration is founded on the premise that the client has sufficient time to take a decision and that he can spend resources to get information. In a low trust culture, people are skeptical about finding quality products if they spend resources so that they relate product selection not to a CKR evaluation, but to luck, to the first information obtained, which is not really connected to the relation between A, B and C. Therefore, from different reasons, both in high trust cultures and in low trust cultures, the application space of CKR is reduced.
Moreover, we live in a society of seduction. In Hirschman’s words, the competitors that cannon maintain competition with technological inputs will not give up competition, but they will use efficient marketing strategies to attract clients. In this way, the clients that are not satisfied with a product will move to another one, and the others will chose the first product, in this way giving the possibility to the low quality products companies to survive. People will make cost-quality CKR choices, but these choices are not very convincing because the information they made the choices on is not entirely valid. This reasoning has one objection, namely that, hypothetically, behind a decision there is information that can lead to cost-benefit type of thinking. We can answer this by saying that seldom the person that takes a decision analyzes the validity of the information and, usually, he takes the decision based on uncertainty and not on risk.
If we take in account the third’s ability to intervene and to sanction betrayal, we find ourselves in front of some problems related to the third’s independence of the other parts involved, to his ability to identify proofs of betrayal, to sanction and implementing the sanction related costs, and all these problems emerge one way or another in the public image. If inside a society cheating became a norm and was associated to state’s inability to sanction, immorality being accompanied by the lack of legality, would we still be able to talk about social order? Yes, we can, but in a totally different way than the one suggested by game theory, whose assumptions are, undoubtedly, ethical. If we add another aspect, namely that the judicial “comes after”, that between actions and their judicial framing gaps may appear, that this framing is focused on a “past” permanently surpassed by “the present”, adding the necessary time to actualize and implement, it will appear as obvious that we cannot rely in our understanding of the new social order on the game theory assumptions.
There are situations when decisions are not based on the logic of the analyzed object, but on the logic of the situation, as we can see in the “rabbit or stag” game:
Table 2
A°
| |||
stag
|
rabbit
| ||
*B
|
stag
|
*3, 3°
|
1 ,0
|
rabbit
|
0, 1
|
*2, 2°
| |
As we can see in the matrix above, the Nash equilibrium are (stag, stag) and (rabbit, rabbit). A hunter that goes hunting with other hunters will win if he hunts what the 2/3 of the participants hunt, and he will lose if he positions himself among the other 1/3. The game illustrates the spontaneous emerging of a social order, based on a majority sustained convention.
Usually, when we have to take a decision, we focus more on the object that needs to be evaluated and less on the way the others act regarding the same problem. Nevertheless, let’s assume that the player has to act “differently”. To succeed, he will have to evaluate the way every other player acts, to estimate their possible decision and to make a quick calculation, similar to a bridge player. In common sense terms, it’s about knowing and acting any way the wind blows. This situation is possible when the meetings are scheduled ahead, when the agenda is known, when the players know each other and when information about the players’ interests is available. Moreover, as North says, we should hold relevant information, and we shouldn’t be concerned about the costs related to obtaining relevant information about the other players’ interests, so that the players’ opinions correspond with their individual actions. In other words, people should not consider expensive promoting their points of view, case in which the relation between opinion and behavior becomes inconsistent by distorting one of the terms of the relation in public space, “we say one thing and we do another.”
Therefore, cultural elements will always influence decisions, by affecting the validity of the information that the people that are taking decisions have. If that culture is a freedom oriented one and the citizens are willing to pay the price of defending their opinions, then it is possible that in a situation that requires a decision both sides to have relevant information and to have more chances of spontaneous arrangements by the contract they made. But if the cultural orientation is toward equality and the costs of expression are high, we can expect to a process of distorting the opinions, which diminishes the quantity of valid information at the hand of the one that takes the decision, which leads to the need of a higher instance – institutional-organizational or state – that can assure the contract.
In what context will people choose to hunt rabbits (less favorable choice) and not stags (more favorable choice)? In what context the majority’s orientation represents an element that should be taken in account? Is it possible that the majority to accept a less favorable convention instead of a more favorable choice, but with fewer supporters if, for example, there were some animal species protection related pressure? In this case, the already given exterior elements, such as the participants’ ecological orientation, will influence the decision and, implicitly, the obtained social order. There is a possibility that this reason to get an answer from the supporters of the game theory, namely that, in that case, not respecting a norm is associated by the players with a utility loss. But if the ecological orientation is a private issue, not confessed by the players, how is it possible that in this case a CAB requirement to function?
The maneuver space for CKR and CAB is also reduced because it assumes the localization principle, meaning that the players are located in the same space and there is a high probability that they will meet again, fact already known by the hunters. The delocalization of space produced by the modern instantaneity reduces the chances that the meetings happen using the same network of players and, in this case, the rule of majority will not work. In other words, the game we talk works when the players are space anchored and under its control.
The game below is also a static one, where the information that each player has about the other ones is important, with the only difference that here the action of “the majority” is not taken into account. The equilibrium solutions are (hawk, dove) and (dove, hawk).
Table 3
A°
| |||
hawk
|
dove
| ||
*B
|
hawk
|
- 4, - 4
|
*3, 2°
|
dove
|
*2, 3°
|
1, 1
| |
These solutions assume the existence of an “active memory” inside the community and of a dependency on this memory. Player A will play “hawk” if he knows that player B will play “dove”. But for the letter to do this, we assume that he is hesitant towards confrontation, information that player A has it. If player B knows that A knows how he acts in a confrontation situation, it is possible that he will modify his behavior, and in this case we ask if A knows that B knows what A knows, etc. To sum it up, the game is played by players that depend one of each other and where reciprocal information and dignity related issues are relevant. How high is the degree of dependence between the two players, in a world in which relations are quickly done and undone? If we talk about “dignity” and “reputation”, the recommendations generated by a solidarity ethics are oriented towards cooperation (win-win strategies). But the “competition” orientation and the idea that this represents the key to succeed determines the players to adopt from the start the “hawk” strategy, without taking into account the way the other one reacts. Moreover, the Nash equilibria in this game assume a “solid” construct of identity (I refer here to Bauman’s description of identity) and of personality too, a “given” that is never negotiated and that is dependent of the past, the past itself being a form of “solid”. In the same way as in the “rabbit or stag” game, the present one will offer a valid understanding of reality if we are localized and if the space controls the players’ behavior using the concept of “credibility”. It is not necessary for the two conditions to be simultaneously realized. An example in this case could be the analysis of the players in the movie “Duel” where the excess of self esteem leads to a “hawk-hawk” type of strategy, where the space is not limitative.
Let’s take into discussion the time dimension and its association with the success of the action. The idea that success comes in time is not new, Tocqueville asserted it when he said that governments should teach people that success comes after long efforts – the theme of “the delayed reward”, defining for the first modernity. If we translated this request in the terms of contracts between two partners, A and B, it would take the form of a dynamic game, such as the one below:
1 2 3 4 5 6 7 8 9 10
|-------|-------|------|-------|------|-------|-------|-------|------|
A: 25 > 20 70 > 40 85 > 70 102 >100 110
B: 10 30 > 20 50 > 40 90 > 80 102 > 101
Looking at the diagram, we ca see that the two partners can come, after repeated contracts in which each one has the advantage on his side, to big gains. The success consist in satisfying one condition, namely that none of the players will take advantage of the information they have and that they will give up this advantage conferred by the information in favor of the other partner.
In reality, in every contract, the parties do not have the same type of information. The differentiate exposure to information determines that one player to be in the position to exploit the situation in his favor, the way that North says it. What would determine him not to? Knowing that he could meet the other player and that he could play an n round game, where both players could win, or the threat of public exposure of the cheating act, which would damage the public image of the one that betrayed. In both cases, there is a hidden assumption, the one of conditioning the players’ behavior.
Moreover, in the game above, there is an assumption which is the key element of a debate in game theory, related to the way of which the players refer to an instrumental rationality or a general one. The solution of the game consists in a translocal regulation. This could happen inside family, where the logic of cooperation is translated in reciprocal surrenders. But does it work in other social situations where one cannot sanction morally or legally the abandonment? The answer could be affirmative with one condition: the abandonment of the concept of instrumental rationality in favor of a type of rationality conceived more as an expression of what it is possible: “it becomes a purpose with its own right” (Hargreaves Heap-Varoufakis). Question: how valid it is, in today’s modernity, an address on general rationality? Can we accept the idea of a fundamental social critic and of the possibility of changing the fundamental sentences throughout such a critic? To answer affirmatively means to accept two requirements – that there is an instance that knows how things are done and “what needs to be done”. Tocqueville, the rightful heir of Montesquieu, considered that this instance is the government; otherwise his book, “The Old Regime and the Revolution”, is nothing else but a description of this status. Marx and then Lenin considered that this instance should be the communist party. Who could assert today this place and the competences required? Do we still have the great integrators and the great motivators?
The game above raises another problem. In the way the game is described, we can say that it starts at the moment 1. In reality we rarely find ourselves in a real virtual moment 1. Usually, our decision is taken by placing ourselves in a situation laid out through a prior decision. Not always the prior decision is ours, as in the case of “path dependency”. How do we explain the fact that we use alternative current and not continuous one, that we have wide or narrow railways, right side or left side steering wheel? That once a decision taken, it represented the ground for such imposing institutional constructions, so that the exit costs are very high, as in the case of social revolutions and not only. In this case, the individual’s decision is marked by a situational context for which he is not responsible of. At this point, both the postmodernists and the theoreticians of game theory meet. For the former, “the path dependency” does not exist because, they say – and here we quote Bauman – the present does not represent anymore the link between past and future. For the latter, “the path dependency” cannot be taken into account because it nullifies the principle of maximum freedom which is the ground on which game theory is built. The acceptance of the concept would oblige the game theory to accept a requirement that it cannot satisfy, namely that in certain moments a player’s decision was taken under the conditions of a reduced freedom space and in different situations under the conditions of a wider freedom space. The linearity and the constancy of the freedom space in a time unit are “negotiable” inside game theory. For sociologists but for historians too this aspect requires the observance of two methodological demands – accepting the variability of the freedom space, the one that generated the decision and the possible existence of “the path dependency”.
Paradoxically, although the late modernity, the present one focuses on individuality and maximum freedom, game theory, although built on the same principles, can provide valid propositions only in well delimitated cases. Here we are in front of a metatheory problem – understanding the fact that although the pre-requisites of a theory are consonant with the pre-requisites of the social fact, the theory has a reduced validity degree in what concerns understanding the way social order is generated.
Niciun comentariu:
Trimiteți un comentariu