http://broadcast.iu.edu/ceremon/celeb07/index.html

Scroll down to Monday, October 15th

Such an interesting debate that took place during the early 20th century

I. Introduction
     Brian Ellis has argued for Scientific Essentialism, a term originally coined by George Bealer in his paper “The Philosophical Limits of Scientific Essentialism.” In that paper, Bealer raises some concerns against the Kripkean position that some essences of natural kinds can be discovered a posteriori. A common natural kind example of water, the essence of water is being composed of H2O. Necessarily, water must be composed of H2O. The essence of water, H2O, was discovered by the natural sciences. Bealer’s concern is about how useful the Kripkean position is for philosophers. Despite Bealer’s concerns, philosophers like Brian Ellis, Caroline Lierse (1992), John Bigelow (1990), and Alexander Bird (2007) have come to embrace Kripke’s insight and make it one of the central themes of Scientific Essentialism.
     This paper will focus on Ellis’s Scientific Essentialism due to the fact that Ellis restricts any discovery of essences to the empirical sciences. However, the Kripkean position only accepts that some essences can be discovered a posteriori. When looking specifically at Kripke’s argument against the mind-brain identity thesis, Kripke presents the example of pain. The essence of pain is its painfulness or the feeling of being in pain. The essence of pain (according to Kripke) can only be discovered a priori. No empirical work can be done to discover the essence of pain. With respect to the pain example, Ellis is silent yet he clearly has to deny the example. The problem for Ellis can be formulated as follows:
(a) Ellis accepts the Kripkean story about natural kinds.
(b) Ellis requires that essences can only be discovered through the
empirical sciences.
(c) If (1), then some essences are discovered a priori.
(d) If (2), then no essence can be discovered a posteriori.
The consequent of (c) and the consequent of (d) contradict so either (a) or (b) has to be rejected.
     First, how Kripke goes from the essence of water is being composed of H2O is an a posteriori discovery to the essence of pain is an a priori discovery will be presented. Second, a few reasons that Ellis (or someone who agrees with Ellis) might be able to give to accept the Kripkean position but reject the story about the essence of pain.

II. Discovering Essences in Kripke
     What makes it the case that some object P has some essence M? An essence M of object P is something the object P must have at every place, time, or possibility in which object P exists. Such an account of essence can be expressed as a biconditional:

(E) M is the essence of P iff (M is with P whenever and wherever P exists
and M is with P whenever and wherever P could exist).

Let us apply (E) to the example of water. There is a glass of ice water near me at the moment and that water has some set of properties, taking the form of the glass, some of the water is in solid form (the ice), and is current at a cold temperature. I wait some amount of time and the water loses some of those properties, the water will become liquid, the water will take the form of my stomach after I drink it, and the temperature of the water will rise, but the water will still exist.
     What could not happen to the water? At first, it seems trivial that what could not happen to the water is that it not be water. It is essential that water is water. Whenever you come across water, you come across water. Such a claim does not seem particularly interesting or informative at first, but what happens when one recognizes that the natural sciences discovered that water is H2O? “Water is H2O” is an identity proposition and with an identity proposition we can substitute an occurrence of one term with the other term in any proposition and preserve the truth of that proposition. From “It is essential that water is water” and “Water is H2O” we get “It is essential that water is H2O.” Combine that last proposition with (E) and we conclude that at any place, time, or possibility, if we come across any water then we will also come across water. Kripke’s argument for H2O as the essence of water can be formalized as follows:

(W1) (Ax)(Ay)(If (x=y) then  (x=y)) (Leibnitz’s Law)
(W2) Water = H2O (Discovery from the empirical
sciences)
Therefore,  Water = H2O (From (W1) and (W2))

     The above argument relies on the authority of science to give us the essence of water and this is where we see the beginnings of Scientific Essentialism. Ellis clearly agrees with the above argument, but this is not the only essence argument which Kripke gives. Next, we need to look at the argument Kripke gives for heat. Kripke acknowledges a distinction between heat and the sensation of heat. The way heat feels to human beings is different from the actual physical phenomena, heat. Kripke claims that science discovered the essence of heat is molecular motion so we can run a similar argument to the water case:

(H1) (Ax)(Ay)(If (x=y) then (x=y)) (Leibnitz’s Law)
(H2) Heat = Molecular Kinetic Energy (Discovery from the empirical
sciences)
Therefore,  Heat = Molecular Kinetic Energy (From (H1) and (H2))

     What about in the case of pain? Is there some divide between the way pain feels and the physical phenomena of pain? Actually, Kripke maintains that pain is the feeling of pain. Consider the claim “Pain is the firing of C-fibers” and if science tells us that such a claim is true do we accept what they say? Kripke’s answer is, no. Why? Because we already know the essence of pain, pain is the feeling of being in pain. Unlike external things like heat and water, pain is something we have immediate epistemic access to and pain is something that cannot be dissociated with the way it feels. One does not have to go to the empirical sciences (according to Kripke) to discover the essence of pain. With that it in mind we get the following argument:

(P1) (x)(y)(If (x=y) then  (x=y)) (Leibnitz’s Law)
(P2) Pain = feeling of being in pain (A proposition any being
who has been in pain knows to be true)
Therefore,  Pain = feeling of being in pain (From (P1) and (P2))

     When we look at Ellis’s work, we see a clear denial of the last argument, “[M]etaphysical necessities have to be discovered by scientific investigation.” For Kripke, scientific investigation will not tell us anything about pain since we already know what is essential or metaphysically necessary for pain. At the same time, Ellis clearly accepts the Kripkean arguments for water and heat.

“Thirty years ago, many Anglo-American philosophers, and probably most Australian philosophers, believed in contingent identities… water is contingently identical with H2O… [heat] is contingently identical with… molecular kinetic energy… Kripke argued, correctly in my view, that the concept of contingent identity is an oxymoron. If a = b, then this relationship holds necessarily, not contingently.”

     Despite this agreement with Kripke, there is no response to Kripke’s pain example and why pain fails to be identical to the feeling of pain. There is no reason from Ellis why we should say that Kripke was wrong with respect to our infallible access to pain. If pain is identical to the feeling of being in pain then scientific investigation cannot tell us anything about pain. Therefore, not all metaphysical necessities have to be discovered by science.

References

To be added soon

[Again, the "box" is just the universal quantifier]

Kamp, Hans, van Genabith, and Reyle (forthcoming) offer a fairly thorough account of the problems with donkey sentences. What they consider first is the standard method for translating natural languages sentences into first-order predicate logic. The existential quantifier is defined as follows:

(1) $xF Û Ø“x ØF

Consider a simple sentence:

            (2) A delegate arrived.

Translated into FOL we get:

            (3) $x(delegate(x) Ù arrived(x))

(3) is logically equivalent to:

(4) Ø“x Ø(delegate(x) Ù arrived(x))

When (4) is translated into English we get:

            (5) It is not the case that every delegate failed to arrive.

Now consider a sentence with anaphora:

            (6) A delegateⁱ arrived. She_i registered.

Kamp, Hans, van Genabith, and Reyle introduce the superscripts and subscripts in order to keep track which noun matches up with which pronoun.

            If we apply what we did in steps (1)-(5) to (6) we get the following sentence:

            (7) It is not the case that every delegateⁱ failed to arrive. She_i registered.

Intuitively, there is something wrong with (7) as it does not capture what is expressed by (6).

References

Kamp, H., J. van Genabith and U. Reyle, forthcoming. “Discourse Representation Theory”, in Gabbay D. and F. Guenthner, Handbook of Philosophical Logic (second edition), Springer.

Our everyday knowledge-attributing practices, according to the contextualist, are going to have varying standards based on how high the stakes are with respect to the knowledge assertion being true. Intuitions concerning Stewart Cohen’s airport case are some of the contextualists’ evidence for maintaining that their position is more in line with the everyday account of knowledge.

Mary and John are at the L.A. airport contemplating taking a certain flight to New York. They want to know whether the flight has a layover in Chicago. They overhear someone ask a passenger Smith if he knows whether the flight stops in Chicago. Smith looks at the flight itinerary he got from the travel agent and respond, ‘Yes I know — it does stop in Chicago.’ It turns out that Mary and John have a very important business contact they have to make at the Chicago airport. Mary says, ‘How reliable is that itinerary? It could contain a misprint. They could have changed the schedule at the last minute.’ Mary and John agree that Smith doesn’t really know that the plane will stop in Chicago. They decide to check with the airline agent. (Cohen 1999 p. 58 )

According to Cohen, if we consider some response that is insensitive with respect to context then one can’t make sense of the above scenario.

The reason Cohen gives for such a conditional is that with any insensitive response, there are only three possible responses. The first is that the standards of Mary and John are too high and Smith does know that the plane will stop in Chicago. The second is that the standards of Smith are too low so Mary and John are right to deny him knowledge. The third is that the standards of Smith, Mary, and John are too low therefore none of them have knowledge. According to Cohen, “None of these answers seem satisfactory.” (Ibid p. 59) Cohen asserts that “…the best answer: Neither standard is simply correct or simply incorrect. Since the standards for knowledge vary across contexts, each claim, Smith’s as well as, Mary and John’s can be correct in the context in which it was made.” (Ibid p. 59)

The reason that Mary and John correctly deny knowledge to Smith is because they are in a high-stakes situation as it is important for them to get to Chicago. This high-stakes situation means their standards for knowledge are higher. With respect to Smith, he is in a low-stakes situation as it is not that important that he get to Chicago. This low-stakes situation means his standards for knowledge are lower than Mary and John’s. In high standards of knowledge the possibility of error becomes salient and so cannot be ignored. Therefore, Mary and John must acquire more evidence so they can know the plane will land in Chicago.

Contextualists maintain that their answer to the above case is more in line with the everyday account of knowledge. They claim their account of the airport case is the everyday use of knowledge. Contextualists recognize it is essential for their view to conform to the everyday account of knowledge otherwise their position is untenable. DeRose is very direct about this when he writes, “…the contextualist’s appeal to varying standards for knowledge in his solution to skepticism would rightly seem unmotivated and ad hoc if we didn’t have independent reason from non-philosophical talk to think such shifts in the content of knowledge attributions occur,” (DeRose 2002 p. 169)

and

“…the best grounds for accepting contextualism concerning knowledge attributions come from how knowledge-attributing (and knowledge-denying) sentences are used in ordinary, non-philosophical talk: What ordinary speakers will count as ‘knowledge’ in some non-philosophical contexts they will deny is such in others.” (DeRose 2005 p. 172)

It seems fairly clear that if contextualism fails to conform to the everyday account of knowledge then contextualism also fails to be a tenable position.

References

Cohen, S., 1999, “Contextualism, Skepticism, and The Structure of Reasons”,

Philosophical Perspectives

DeRose (1992) “Contextualism and Knowledge Attributions”, Philosophy and

Phenomenological Research

(2005), “The Ordinary Language Basis for Contextualism and the New

Invariantism”, The Philosophical Quarterly

[I don't know why, but the "all" quantifier is coming out as a box. I have seen that before, but usually the "some" quantifer also comes out as a box which it isn't in this case. Just read the box as an "all" quantifier]

Well, it has been a while since I posted. I think I am still recovering from Spring 2008 semester. Well, whatever the reason (or excuse) is, I hope to start posting again. I have been trying to work out the problem known as “Donkey sentences.” I realize when discussing the topic with some of my colleagues, they have trouble trying to figure out what the problem. Therefore, I have been trying to more clearly state the problem than the readings I have come across.

The problem of what have come to be called “Donkey sentences” has to do with a certain kind of anaphora (statements about other statements) has received considerable attention since it was reintroduced in Peter Geach’s Reference and Generality. The problem is often introduced as follows, consider the following sentences:

(1) If Pedro owns a donkey then he beats it.

(2) Every farmer who owns a donkey beats it.

What has taken to be the “natural” strong reading of those sentences is: (1) is the case only if Pedro beats all the donkeys that he owns and (2) is the case only if every farmer beats the donkey(s) she/he owns. The natural readings of (1) and (2) has created considerable problems when trying to translate them into First-Order-Logic (FOL) as no FOL translation can capture these natural readings. There have been two important responses to this problem, Discourse Representation Theory (DRT) proposed by Hans Kamp in 1981 and Independence Friendly Logic (IF Logic) proposed by Jaakko Hintikka in 1985. Both solutions have had little discussion between the two of them and so both have developed rather independently of each other.[1]

Since both offer different solutions to the same problem it should be determined which (if either) side offers a better solution to the problem. After reviewing both sides I will argue that IF Logic has a better solution to the problem because it is more in line with FOL and so is a simpler and less ad hoc solution. My strategy for approaching this problem will be as follows; first I will sketch out a more precise account of the problem. Second, I will summarize DRT’s solution and IF Logic’s solution. Third, I will give my reasons for claiming that IF is a simpler and less ad hoc solution.

II. The Problem with Donkey Sentences

There have been attempts to translate sentences (1) and (2) into FOL that can capture the truth conditions already mentioned. With respect to (1), three translations have been offered:

(1a) $x[donkey(x) Ù (owns(pedro, x) ® beats(pedro, x))]

(1b) $x[(donkey(x) Ù owns(pedro, x))] ® beats(pedro, x)

(1c) “x[(donkey(x) Ù owns(pedro, x)) ® beats(pedro, x)]

Similar translations have been attempted with respect to (2),

(2a) “x[(farmer (x) Ù $y(donkey(y) Ù owns(x, y))) ® beats(x, y)]

(2b) “x$y[(farmer (x) Ù donkey(y) Ù (owns(x, y) ® beats(x, y))]

(2c) “x“y[(farmer (x) Ù donkey(y) Ù owns(x, y)) ® beats(x, y)]

The problem with (1b) is fairly obvious as it does not express a well-formed formula (wff) because it leaves a free occurrence of the bound variable x. Why the remaining propositions fail needs a little more explanation.

             The reason (1a) fails is the proposition comes out true whenever there is a donkey that Pedro doesn’t happen to own. Demonstrating that (1a) comes out true in such an instance is fairly simple given the definition of the conditional and conjunction. The conditional comes out false if and only if (iff) the antecedent is true and the consequence false and true otherwise. The definition of conjunction says that it will come out true iff both conjuncts are true and false otherwise. Suppose that x is a donkey, but it is a donkey that is not owned by Pedro. Such a state of affairs means the conditional in (1a) will come out true as the antecedent (owns(pedro, x)) will be false therefore the conjunct (owns(pedro, x) ® beats(pedro, x)) comes out true. Since x is a donkey then the other conjunct (donkey(x)) also comes out true. Therefore both conjuncts are true; therefore the proposition is true. This fails to capture the natural reading of (1) as (1) is supposed to only come out true when Pedro beats all the donkeys he owns.

            The reason (1c) fails is the proposition comes out true when Pedro owns something other than a donkey. Suppose that Pedro owns a pig and ignore whether he beats it or not. Such a state of affairs means that one of the conjuncts in the antecedent is false (donkey(x)). Therefore, the conjunction (donkey(x) Ù owns(pedro, x)) is also false. Therefore, the conditional is true as it has a false antecedent. If it is the case that Pedro can own a pig and this ownership will make (1c) come out true then clearly this fails to capture the natural reading of (1).  

              Scientific Essentialism is a view that has received increased attention within recent works[1]. Two central features of this view are:

(i) “…inanimate matter is not passive, but essentially active,” and

(ii) “…anything that belongs to a natural kind is logically required (or is necessarily disposed) to behave as its essential properties dictate.”[2]

What Ellis takes to be the opposing view is the “Humean” position which has as one of its central features being:

(H)“…things behave as they are required to by the laws of nature.”[3]

The question can easily be raised, if matter is active, its activity is the result of its essential properties, and its activity is not required by the laws of nature, then what reason do we have to think that there are laws of nature when activity comes from matter? Stephen Mumford[4] has answered such a question in the negative. Mumford points out many problems with essentialism, but his rejection of laws seems (as I will argue) more in line with scientific essentialism.

Another position that scientific essentialism is committed to is scientific realism.[5] Scientific realism, according Richard Boyd, maintains the “…product of successful scientific research is knowledge of largely theory-independent phenomena and that such knowledge is possible (indeed actual).” [6] It has been argued that a significant number of scientific theories have been acquired through the application of simplicity. If theory T1 is simpler than theory T2 then we should choose theory T1.The scientific realist (essentialist) can easily ask the following questions, if a theory is chosen because it is simple then does the theory have anything to do with the theory-independent phenomena?

            If such a question does not have an acceptable answer for the scientific essentialist then the scientific essentialist must reject simplicity as a characteristic of a good theory. This creates a dilemma for the scientific essentialist, (1) Thomas Kuhn[7] maintained that, historically, some scientific theories have been accepted because they were simpler than the alternatives; (2) Hugh Gauch’s[8] account of scientific practices maintains many scientific theories apply simplicity when dealing with the “curve-fitting problem.” Mario Bunge[9] and Richard Boyd[10] have argued that the historic importance of simplicity has been exaggerated by philosophers of science and really has little or no importance when choosing between scientific theories.

            What remains is the “curve-fitting problem.” Elliot Sober[11] has produced a number of works attempting to maintain both scientific realism and justify simplicity within such a doctrine. After reviewing the various works concerning these topics, I have concluded that simplicity is inconsistent with scientific essentialism by an argument that will be presented within this paper. If such an argument is correct then I foresee two options either another account of simplicity needs to be given or simplicity needs to be abandoned for some other notion within scientific essentialism. While I leave the former as a possibility, my focus will be on the latter. The dispositional account of science suggested by Mumford[12] gives a better interpretation of scientific essentialism.

If theory T1 is simpler than theory T2 then, all other things being equal, we should choose theory T1. Such a conditional, according to Alan Baker[1], most philosophers would accept. This has created a problem for those philosophers that accept the doctrine of scientific realism, whether or not to reject the notion of simplicity as a characteristic of a good theory. Scientific realism, according Richard Boyd[2], maintains the “…product of successful scientific research is knowledge of largely theory-independent phenomena and that such knowledge is possible (indeed actual).” The scientific realist can easily ask the following questions, if a theory is chosen because it is simple then does the theory have anything to do with the theory-independent phenomena? Why think that the simpler theory is more likely to give us knowledge of these theory-independent phenomena?

            If such questions do not have an acceptable answer for the scientific realist then the scientific realist must reject simplicity as a characteristic of a good theory. This creates a dilemma for the scientific realist, (1) Thomas Kuhn[3] maintained that, historically, some scientific theories have been accepted because they were simpler than the alternatives; (2) Hugh Gauch’s[4] account of scientific practices maintains many scientific theories apply simplicity when dealing with the “curve-fitting problem.” If the scientific realist rejects simplicity then either she needs to maintain all of the theories where simplicity was applied as unjustified or give some other interpretation as to why these theories are correct. Mario Bunge[5] and Richard Boyd[6] have argued that the historic importance of simplicity has been exaggerated by philosophers of science and really has little or no importance when choosing between scientific theories.

            What remains is the “curve-fitting problem.” For over the past 30 years, Elliot Sober[7] has produced a number of works attempting to maintain both scientific realism and justify simplicity within such a doctrine. After reviewing the various works concerning these topics, I have concluded that simplicity is inconsistent with scientific realism by an argument that will be presented within this paper. If such an argument is correct then I foresee two options either another account of simplicity needs to be given or simplicity needs to be abandoned for some other notion. While I leave the former as a possibility, my focus will be on the latter. The dispositional account of science suggested by Mumford[8] gives a better interpretation for the scientific realist.

Now that was one busy semester!!! If I have to read Kant within this lifetime, it will be too soon. But I did survive, I was scared at one point that I wouldn’t. Formal Logic, Kant, and Theory of Knowledge are some of the more labor intensive courses at UNL. The semester kept me from updating my blog, but I am glad to be posting again.

I enjoyed the Formal Logic class. I am concerned about this focus towards the meta stuff though. I find what seems to be going on within the language more interesting than the meta stuff. I am glad that I can do proofs within paraconsistent and intuitionistic logic. I want to get better with Fuzzy Logic though, that is a very interesting logic. I hope to get some practice over the summer.

Kit Fine gave a very cool presentation at UNL this semester on vagueness. He has a proof which he thinks shows that given classical logic, any account of vagueness will be impossible. He is willing to give up on some aspects of classical logic in order to make progress on the problem. Giving up on classical logic concerns me (though I am not sure as to why, I am worried my support for classical logic is not well thought out). I think it was one of the few times I agreed with Dr. Becker. He maintained that if Fine is right then we have two choices, give up on some aspects of classical logic or just pick some arbitrary point and say person 53 is bald and person 54 isn’t. I am certainly not opposed to the idea, but there is something a little too easy about such an answer. While that isn’t a reason to reject these arbitrary choices, it gives me the “intuition” that something is wrong.

I hope this will be a productive summer (though I have mainly been recovering this past week). I want to significantly improve my paper on dispositional essentialism, but it needs a lot of work. Same with my paper on simplicity. I want to read through Pritchard’s new book “Epistemic Luck,” Salmon’s book “Reference and Essence,” Cartwright’s “The Dappled World,” and more closely Mumford’s “Laws in Nature.”

Alvin Plantinga[1] proposed an account of justification (he prefers the word “warrant”) which he believes to be the correct account. Responses to his account have produced several Gettier type counter-examples which appear to defeat his account. Plantinga[2] conceded these examples by purposing several amendments to his definition of warrant in order to get around these counter-examples. Thomas Crisp[3] argued there are problems with these amendments which make them incapable of getting around the Gettier type counter-examples they were meant to deal with. Crisp also proposes his own amendments to Plantinga’s account of warrant which he believes works better when dealing with Gettier-style counter-examples. I argue that Crisp’s proposed amendments also fails given another Gettier type counter-example.

This paper has three parts. First, a very brief summary of Plantinga’s account of warrant and the scenario that creates a problem for his account. Next will be Crisp’s critique of Plantinga’s solution and his own solution to the problem. Finally, I will offer a scenario that is a counter-example to Crisp’s amendments.

II. Relevant parts of Plantinga’s account of Warrant and where it fails

Plantinga argues that warrant comes in degrees. One can have some warrant for false beliefs, but if an agent has enough warrant then one has knowledge. This differs from something like a view of justification that allows for an agent to have a lot of justification for a belief, but if the belief is false then it isn’t knowledge. Plantinga rejects this notion of justification; an agent cannot have enough warrant for knowledge and not have knowledge. In order for the agent to gain enough warrant for knowledge it has to be done by properly functioning processes of the epistemic agent within a favorable cognitive environment. For the purposes of this paper, the only important part is the “favorable cognitive environment.”

            There are Gettier-style scenarios where Plantinga’s account of warrant is unsuccessful at getting the desired result. Consider the following scenario[4]: Jack attends a Fighting Irish football game. Jack arrives at the game in his Chevrolet van. Jack parks in one of the football coaches’ spots. The players, not wanting to let this act go unpunished arrange to have Jack’s truck towed from the parking lot and crushed into a cube shortly after he leaves his vehicle. Unbeknownst to Jack, he has won a contest which will be announced during the game and the prize is a Chevrolet van. He spots a friend on the way to his seat and his friend happens to ask him what vehicle he owns. Jack responds, “I own a Chevrolet van.” Such a belief is true, but it is only accidently true. Consider a very similar counter-factual scenario where the players tow the van, but don’t destroy it. Such a counterfactual intuitively appears to maintained knowledge for Jack.

            Given Plantinga’s account of warrant, either both scenarios are instances of knowledge or neither are instances of knowledge because the

“…belief is produced by the very same cognitive processes functioning the very same way in the same favorable cognitive environment. Hence, … either both of these situations in which [Jack] know[s] that [Jack] own[s] a Chevrolet van, or neither is.”[5]

Obviously, Plantinga wants to be able to reject this conclusion within his account of warrant. There doesn’t seem to be anything wrong with the agent’s cognitive processes in either scenario so Plantinga rejects the idea that the environment is cognitively favorable.

            In order to avoid the destroyed van scenario, Plantinga introduces the notion of a “mini-environment” and offers a new definition of “favorability”.

Plantinga defines “mini-environment” (ME) as:

(CME) a state of affairs A is a mini-environment with respect to exercise E of cognitive power in maximally specific epistemic circumstances C =_df (i) A is properly included in C, and (ii) A is as much as possible like C given that A entails neither the proposition that E yields true belief nor its denial.[6]

Plantinga defines “favorable” as:

(F) a cognitive mini-environment is favourable with respect to a particular exercise E of S’s cognitive powers if and only if, if S were to form a belief by way of E in this mini-environment.[7]

Given (CME) and (F) Plantinga concludes that:

(RC) A belief B produced by an exercise E of cognitive powers in a cognitive mini-environment ME has warrant (sufficient for knowledge) only if ME is favourable for E.[8]

Plantinga maintains that (CME), (F), and (RC) will allow his account of warrant to avoid the destroyed van scenario. The destroyed van scenario is an ME that is unfavorable for E therefore the belief B is not warranted and so does not qualify as knowledge.

III. Crisp’s definition of Favorable Mini-environments

            Crisp argues that Plantinga’s amendments will not get the desired result. Crisp notices within the destroyed van scenario there are two issues and only one of these issues is addressed by (CME), (F), and (RC). The first issue is that Jack’s belief lacks a kind of general favorability because it is an accidently true belief. The other issue is that the scenario is very close to a counterfactual possible world where Jack has knowledge so Jack’s belief also lacks counterfactual favorability. Plantinga’s amendments only focus on the counterfactual favorability. Applying (CME), (F), and (RC) produce the undesirable result; if a scenario is counterfactually favorable then that scenario will not be accidently true.

            In order to illustrate the above point, Crisp offers a scenario that is counterfactually favorable but still an accidently true belief. Consider the scenario where a small town has a monthly, “Guess the number of prunes in the jar” contest. You are a citizen of this town and your uncle runs the contest. You have a friend and he tells you that there are 138 prunes in the jar this month. He got the information from someone who works for your uncle but the information is faulty. Your uncle comes down with an odd brain disease which causes him to think there will be a major disaster if you don’t win this months contest. What your uncle does is no matter what number you write down, he will empty the jar of prunes and place that number of prunes in the jar. Your belief is true, but it is true by accident. Your belief is a counterfactually favorable because no matter the number of prunes you guess are in the jar, your uncle will put that number of prunes in the jar.

            In order to deal with the prune scenario, Crisp offers the following concepts and a definition for an unfavorable mini-environment:

…‘P(p/q)’ short for ‘the epistemic probability of p given q.

…q confirms p for some S iff P(p/q & k) > (p/k), where k is the conjunction of S’s background beliefs. And say that a proposition p defeats a belief b for S iff P(b/ p & k) < P(b/k) and P(b/ p & k) < n, where n is some real number representing the point at which a human being  with properly functioning faculties would cease to believe that b. Finally, say that p is an undefeated defeator of b for S iff p defeats b for S and there is no true proposition q such that (i) q defeats neither b nor not-b for S, and (ii) P(b/ p & q & k) > n.

(U) ME is an unfavorable mini-environment for exercise E of S’s cognitive power resulting in the belief that b =_df there are propositions p and q, neither of which are believed by S, such that

                        (a) ME entails not-p and q, and

                        (b) P(p/k) > P(q/k), and

                        (c) both p and q confirm b for S, and

(d) not-p is an undefeated defeator of b for S and not-q does not defeat b for S.[9]

These definitions give Crisp the desired result that the destroyed van scenario and the prune scenario are unfavorable scenarios.

b =       ‘I (Jones) owns a Chevy van.’

p =       ‘It is not the case that my van has been destroyed.’

q =       ‘I won the van contest’[10]

and

b =       ‘The jar contains 138 prunes’.

p =       ‘It is not the case that the number of prunes has changed since you submitted your contest entry.’

q =       ‘Your uncle placed 138 prunes in the jar because you guessed there were 138 prunes’[11]

Given these desirable results, Crisp goes on to define favorability with respect to the definition of unfavorability.

(F¢) ME is favorable for exercise E of S’s cognitive powers =_df ME is not unfavorable for E.[12]

IV. A Counter-example to Crisp’s definition of favorability

            Let us consider a slightly modified scenario that has been the source of discussion for many logicians[13]. We have a student, Jane, who is a logic student. On a Monday the professor makes the following announcement to the class, “There will be an exam this week and you will be surprised on the day we take it.” Monday night, Jane decides to work out the implications of what the professor just said.

She supposes the test might be on Friday. If the test has not been given by Thursday then she will know that the test will be on Friday. Hence she cannot be surprised if the test is on Friday. Hence the test cannot be on Friday since she will not be surprised. She has ruled out the possibility that the test is going to be on Friday. The same steps can be used for the remaining days of the week so she concludes there is no way the professor uttered a true proposition. Having ruled out Friday she considers Thursday. If they get till Wednesday and she hasn’t received the test then she will not be surprised that it is on Thursday. Hence, the test cannot be on Thursday. Such reasoning allows her to rule out all the days of the week so she forms the belief that there will be no test that week.

The professor did not realize the logical implications of her utterance and she decides to administer the test on Wednesday. Unfortunately, the professor catches a serious flu Monday night and is not able to prepare a test for any day that week. Jane’s belief is true, but it is purely by accident. Had the professor not caught the flu, her belief would have been false. Consider this second scenario: There is a close possible world where the professor happens to check the logical implications of her utterance and realized she should not give them a test that week. She still gets the flu but since she was not going to give the students a test anyways, Jane’s belief seems to count as knowledge.

            With the following scenario let us consider Crisp’s definition of an unfavorable mini-environment. Our b, p, and q will be as follows:

b = I (Jane) will not have a test this week.

p = Jane’s professor did not check the logical implications of her utterance.

q = The professor caught a serious flu Monday night.

We will only need to be concerned with Crisp’s condition (a) (ME entails not-p and q). Since the mini-environment of the first scenario entails p it does not satisfy the (a) condition. Since condition (a) is not satisfied then the definition (U) is not satisfied. Given the fact that (U) is not satisfied then given the definition of (F¢) we can conclude that the ME in the first scenario is a favorable mini-environment yet the original problem is still present. In both scenarios, Jane’s belief is produced by the very same processes functioning the very same way in the same favorable cognitive environment. The belief from the first and second scenarios has to count as knowledge or neither can count as knowledge, given Plantinga’s account of warrant. Also, Crisp’s concepts and definitions fail to get the desired result within the first scenario since Jane has the true belief that is only true by accident and yet satisfies his favorability requirement.

References

Crisp, Thomas M. “Gettier and Plantinga’s revised account of warrant.” Analysis Volume

60, Number 1, January 2000, p. 42-50

Gettier, Edmund. “Is Justified True Belief Knowledge?” Analysis 23 1963, p. 121-23

Plantinga, Alvin. Warrant: The Current Debate. Oxford University Press. Oxford, UK 

1993a

Plantinga, Alvin. Warrant: The Proper Function. Oxford University Press. Oxford, UK 

1993b

 Plantinga, Alvin. “Respondeo.” In Warrant in Contemporary Epistemology: Essays in

Honor of Plantinga’s Theory of Knowledge. Jonathan L. Kvanvig (Editor). Rowman & Littlefield Publishers, Inc. Lanham, Maryland  1996 p. 307-378

Plantinga, Alvin. “Warrant and Accidentally True Belief.” Analysis Volume 57, Number

2 April 1997, p. 140-145

 Smullyan, Raymond 1987. Forever Undecided. New York: Alfred A. Knopf Press NY  

Given OR and assuming entities either exist or they don’t exist there is no circumstance where two theories are equally accurate and differ in terms of simplicity. In order to demonstrate this let us suppose that some theory, T1 is more complicated than another theory, T2. The only way for T1 to be more complicated than T2 (given OR) is for T1 to postulate more entities than T2. This means that T1 says that the entities not mentioned within T2 are present within the world. The number of entities that they postulate has to be different in order for one theory to be more or less complicated than another.

Let us consider the most basic case possible where one theory is more complicated than another, T1 postulates only 1 entity and T2 postulates 0 entities. This means T1 is more complicated than T2. How could T1 and T2 be equally accurate? T1 postulates only 1 entity and in order for T1 to be accurate that entity has to exist. T2 postulates no entities and so if T1 is accurate then there is no way that T2 can also be accurate (since the world would contain only 1 entity and no entities). If T2 is accurate then there is no way for T1 to be accurate.

Considering the next possible case, T1 postulates only 2 entities and T2 postulates only 1 entity. The same situation is present, T1 postulates only 2 entities and either those entities exist or they don’t. T2 postulates only 1 entity and so if T1 is accurate then there is no way that T2 can also be accurate. If T2 is accurate then there is no way for T1 to also be accurate. This can be shown no matter how many entities are postulated in T1 or T2. It is not possible for two theories to differ in terms of simplicity and be equal in terms of accuracy.