Author Archive


how to screen off evolutionary game theory: which problems are solvable and which are not?

In Uncategorized on April 22, 2010 by wenwenfan

I had this question since Lynn asked about the Odd Man Out game in the seminar. I talked to Sheng about it, and we had some tentative answers, but I don’t know what other people think. I would like to hear from you about it.

In Yasha’s paper, he realized a problem for the Odd Man out game to model coalitions: the game predicts a much higher frequency of the break-up of coalitions. This problem can be interpreted in two ways: either it does not fulfill the empirical criterion (because its prediction is disconfirmed by emprical data) or there is some feature the game should incorporate. Yasha takes the second interpretation. This case seems the way Zac does in his paper about the Prisoner’s Dimma game. Zac found out that the Prisoner’s Dilemma game cannot yield a prediction of altruistic behavior, so he identified two features (namely, iteration and correlation) that should be incorporated into the game. As long as those features are considered, the game can yield a good prediction.

My question is: how do we know when the model simply fails to meet the empirical criterion and when it neglects some essential features it should consider? Sheng and I thought that perhaps we need some standard to judge when the empirical criterion is met and when it’s not. It may also boil down to the question what features a game should represent what it should not.


the relationship between natural selection and niche construction

In Uncategorized on April 13, 2010 by wenwenfan

I am curious about the relationship between natural selection and niche construction. It seems truism, as Odling-Smee et al. propose, that they are both processes of evolution and that neither process is independent. If so, how do we characterize their relationship? It is not exactly the relationship between chicken and egg, for chicken and egg appear one after another rather than together. Nor is it exactly the relationship between an action and a reaction described in Newtonian physics, for action and reaction cancels each other out but natural selection and niche construction do not.

What do you think?


a question about Walsh’s “Variance, Invariance and Statistical Explanation”

In Uncategorized on April 7, 2010 by wenwenfan

I have a question about the independence of each domain of scientific explanations.

Walsh argues that there are various distinct, autonomous modes of scientific explanations. How do we understand different domains of explanation? It seems to me that a causal explanation can be statistical or mathematical, but that means that the domains are not independent of each other.


thoughts on Pincock’s paper and Saatsi’s paper

In Uncategorized on March 17, 2010 by wenwenfan

Andre asked me to post my questions on the readings. I am afraid it will be a long post in the end, but it should be easy to follow. The words in paratheses are my comments.

Pincock, “Mathematical Contributions to Scientific Explanation”
1. Pincock prefers his revised explanatory indispensability argument (EIA) to Baker’s. He acknowledges that his version begs the question. I can see that Baker’s does as well (talking about “mathematical objects” seems to assume that mathematical objects exist). If they equally beg the question, why should we prefer Pincock’s version?
-Pincock thinks that the real question is about the truth value of mathematical claims, not about whether there exist any mathematical objects. So he thinks that his version of EIA better suits the debate. Furthermore, he thinks that (2′) but not (2) is accepted by the indexing people, which seems to be a merit of his version of EIA. (I don’t think this is an important progress, since the indexing people will still obejct to (1′).)
2. Why does Pincock think that pointing out the ways mathematics contributes to scientific explanations proves P1 (i.e., we ought rationally to believe in the truth of any claim that plays an indispensable explanatory role in our best scientific theories) true?
Since P1 is a very general claim, it doesn’t seem sufficient to be proved by some putative contributions of mathematics. Furthermore, I doubt whether talking about the truth of a claim is intelligible. Do we really think that a weird claim that can explain something, say ET robbed my apartment is an explanation for my losing all the valuables in my apartment, proves the claim to be true?
– By ‘truth’, Pincock means truth values. P1, or (1′) is actually quite weak. It just means that we should (rationally) believe that any claim that plays an indispensable explanatory role in our best scientific theories is true. (1′) is supported by inference to the best explanation. In short, (1′) means that we should believe that the best explanation is true. (Actually pointing out the ways matheatics contributes to scientific explanations does not support (1′), I would say. It is because P1 says nothing about mathematical claims. I would say that those contributions are relevant to (2′) only.)
3. What’s the role of inference to the best explanation (IBE)? What kind of claims can we infer from it?
– The role of IBE for Pincock is to prove the truth values of the best explanation. (I guess he’s talking about a binary value theory, that a claim can only be true or false.)
4. Is the replacement test a good one? Is it biased towards mathematical explanations?
– Pincock advocates this test. He admits that there may be some limitations because we need to keep the explanadum and the crucial premise fixed. He regards an explanation as an argument, where the explanadum is the conclusion and the explanans (in his words, “the things that do the heavy lifting”) as the premises.
As for whether the test is is biased, Pincock won’t say so. But he also seems to admit that there is a question of why the explanadum involves mathematical concept or theorem (as in the cicada case). (But in the case, the replacement test needs justification.)
Saatsi , “Discussion: The Enhanced Indispensability Argument: Representational vs. Explanatory Role of Mathematics in Science”
1. What kind of things are we looking for in an explanation? How deep should the explanation go? (Are the answers objective or subjective?) When can we say that a factor is non-explanatory?
– To Pincock, what we look for in an explanation is everything that helps us explain the explanadum. The explanation cannot stop at a level where we can still keep asking what explains the explanan. It seems that he thinks that how deep an explanation go is objective because there is a fact about it, but it is also subjective in the sense that it depends on the explanadum (e.g., if the explanadum involves mathematical concepts or theorems, then probably the explanation needs to be mathematical).
2. Saatsi says that a map is not explanatory but representational. However, it makes sense to say that a map, which he thinks is merely representational, explains why we can go from building A to building B in a certain way. Pincock talks about three ways for mathematics to be explanatory. Can we say that a representational role and an explanatory role are not exclusive? Why does Saatsi think that they need to be exclusive?
– To Pincock, a representational role and an explanatory role are not exclusive. He thinks that a mathematical claim can be explanatory by virtue of being representational of some recurring phenomenon, like the cicada case. Saatsi thinks that the two roles are exclusive because he thinks that explanations can stop at the level where we can explain the explanadum and that epistemolgoical claims for the explanan is not an explanation. (If Pincock is right, then Putnam’s peg/board case cannot stop at macro-level theories and must appeal to micro-level theories. I would say that where we can stop depends on the explanandum. We don’t always need to go to the deepest level to explain anything.)
3. Does it make sense to say that an explanatory physical fact can be reduced to a mathematical claim so that the mathematical claim is explanatory?
– Pincock accepts mathematical explanations but whether this is reduction is a question. He doesn’t need to accept reduction in order to preserve mathematical explanations. It seems that not endorsing a reduction to mathematical explanations is easier to defend than endorsing it. But maybe in some cases it’s all right to accept reduction. So, I guess he will leave it open.


very basic questions (2)

In Uncategorized on March 4, 2010 by wenwenfan

My last post will reach 200-word limit very soon, so I post a new question here.

2. explanation/confirmation (relative to optimalit models)

When we discussed optimalit models on Tuesday, it seems that how to use the models hinges on the explanation/confirmation distinction. I  get confused about this distinction. What purpose does it serve? (Is it relevant to the explanation/prediction ditinction?)

Collin says that optimality models involve two steps, one is about the equilibrium that represents the input and output and the other is about why/how the input becomes the output (?). I really need someone to explain to me about those steps.

If optimality models are just about the equilibrium, why do we need the second step? How does the explanation/confirmation distinction tie in? How do we know whether an optimalit model is good or not? Can it be used for testing adaptationism (like Sober thinks) or not (like Brandom thinks)? 



very basic questions

In Uncategorized on March 4, 2010 by wenwenfan

Hi all,

My first exposure to philosophy of science is this class, so I really get confused at some very basic distinctions or terms. I hope you can help me out.

1. model vs. theory

Dan told me that a theory has scientific language and can be tested  while a model has not scientific language. If a model cannot be said to be true or false, just whether it can be applied to explain/predict a certain phenomenon, then how can we judge whether a model is good or bad relative to the target phenomenon?

Weisberg talks about several kinds of representational ideals. It seems that if a model can achieve the desired ideal, then it is good. Then, does it mean that whether a model is good is relative to the time as well? For example, when technology advances, we need to de-idealize Galilean idealization. But is it true with the other idealization, that what is a good minimal idealization now is not a good one in the future? So, there is no best model period?