Hintikka s Socratic Epistemology Meets Gettier s Counterexamples

Hintikka s Socratic Epistemology Meets Gettier s Counterexamples John Ian K. Boongaling Abstract The overall goal of this paper is to apply Hintikka s Socratic Epistemology to Gettier s counterexamples to the tripartite definition of knowledge as justified true belief. In the process, I will make full use of Socratic Epistemology s methodology and commitments. This includes, among other things, looking at Gettier s counterexamples as games between an Inquirer and Nature (the source of information), as well as treating the items in them as pieces of information. The strategy that I employ in this paper also makes use of frames (or partitions). One of the important results of this paper is a scenario where Gettier cannot setup the dilemma for the tripartite definition of knowledge. Keywords: Hintikka s socratic epistemology, Gettier s counterexamples, knowledge as justified true belief, interrogative model of inquiry, deductive and interrogative moves, logic of questions and answers 1 Introduction One of the reasons why epistemic agents like you and me take serious interest in knowing that something is the case is because having knowledge is valuable, that such a state is preferable to ignorance. If we can grant this, then perhaps we can also grant something else: that as epistemic agents, we aim (or should aim) at knowledge. While these ideas are unproblematic for most (if not all) of us, some disagreements may soon appear once we get into the details. To begin with, there seems to be no hope of defining the concept of knowledge without dealing with the difficult problems that are associated with the concepts that traditionally comprise it: (1) truth, (2) belief, and (3) justification. As is well known, the idea that knowledge is justified true belief (henceforth JTB) may be traced back to the writings of Plato. For Kriterion Journal of Philosophy, 2017, 31(3): 25 56. http://www.kriterion-journal-of-philosophy.org c 2017 The author

26 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 instance, in 201d of the Theaetetus, Theodorus pupil, Theaetetus, responds to Socrates in the following way: Oh, yes, Socrates, that s just what I once heard a man say; I had forgotten, but now it s coming back to me. He said that it is true judgment with an account that is knowledge; true judgment without an account falls outside of knowledge. And he said that the things of which there is no account are not knowable (yes, he actually called them that), while those which have an account are knowable. [7, p. 223, emphasis added] Subsequent scholars refer to the foregoing as the tripartite definition (or standard analysis) of knowledge. The tripartite definition of knowledge, however, had been the target of what contemporary epistemologists refer to as the Gettier counterexamples (or Gettier cases). In a landmark paper, Edmund Gettier sums up the knowledge as JTB thesis in the following way: a subject S knows that P (where P is a proposition) if and only if the following conditions hold: (1) P is true, (2) S believes that P, and (3) S is justified in believing that P [1, p. 121]. In a very general way, it can be said that the Gettier counterexamples (henceforth GCs) demonstrate that it is possible for all three conditions to be satisfied by an epistemic agent and still fail to have knowledge. For most of us, philosophy involves logical and conceptual analysis, and perhaps, this is the reason why a significant number of philosophers agree that the GCs count as successful refutations of the JTB thesis. In fact, Ernest Sosa, et al states: Edmund Gettier s landmark paper successfully refuted the traditional analysis of knowledge as justified true belief. Through a series of examples, Gettier shows that one can believe what is true and be justified in so believing and yet fail to know. Justified true belief is not sufficient for knowledge. [8, p. 189, (emphasis added)] To further strengthen the point that a significant number of philosophers agree that the GCs count as successful refutations of the JTB thesis, consider Michael Huemer s description of Gettier s overall strategy below: Edmund Gettier famously refuted the justified, true belief theory by means of a type of counter-example now referred to

John Ian K. Boongaling: Hintikka s Socratic Epistemology 27 as a Gettier case. There are two possible kinds of counterexample to a definition. The first kind is one that would show the definition to be too broad; in this case, this would mean an example of something that in fact is not knowledge, but that the definition would count as knowledge... The other kind is that which would show the definition to be too narrow; in this case, an example of something that in fact is knowledge but that the definition would fail to classify as knowledge. Gettier s counter-examples are of the first kind; in other words, they show that justified, true belief is not sufficient for knowledge. [4, p. 436 (emphasis in the original)] From the foregoing discussion, it is safe to say that at this point in time, no philosopher may easily maintain that the JTB thesis is correct in the absence of a successful argument that shows that the GCs are mistaken. In my estimation, this shows that the GCs exert tremendous impact not only on attempts to construe knowledge as JTB but also to other attempts that seek to identify conditions that are individually necessary and jointly sufficient for knowledge. Given these things, one may ask whether or not contemporary epistemology should be concerned with the task of identifying these conditions. In other words, we may ask ourselves whether or not contemporary epistemology should be concerned with the definition (or the analysis) of knowledge. In relation to the questions above, Jaakko Hintikka offers a different view through his Socratic Epistemology (henceforth SE) or what he refers to as the Interrogative Model of Inquiry in his earlier works. 1 In a way, Hintikka s SE, as a model of inquiry, is revolutionary in the sense that it envisions a way of doing epistemology without the concepts of knowledge and belief. Consider what Hintikka says in the following: If we review the questioning process through which we obtain our knowledge and justify it and inventory the concepts employed in the process, we find all the notions of a logic of questions and answers, the notions of ordinary deductive logic, and something like the notions of acceptance and rejection in the form of rules of bracketing and unbracketing. We also find a notion roughly tantamount to the concept of information. What we do not find are philosophers concepts of knowledge and belief. Hence, the problems of knowledge acquisition can be examined, and must be examined, without using the two concepts. This is perhaps not surprising, for if knowledge is going to be the end product of interrogative

28 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 inquiry, it cannot be one of the means of reaching this goal. [2, p. 9 (emphasis added)] At this point, I have only described (in a very general way) the significant items that I would like to investigate in this paper: (1) the tripartite definition of knowledge as JTB, (2) the GCs as refutations of (or at the very least, difficult challenges to) the JTB thesis, and (3) Hintikka s SE and its revolutionary character. In the proceeding parts of the paper, I will discuss these items in a more precise and detailed manner. As I have mentioned earlier, the overall goal of this paper is to apply Hintikka s SE to the GCs. Allow me to briefly explain why I think the aforementioned goal is both important and interesting especially for philosophers interested in contemporary epistemology (and this requires a little bit of context). As is well known, epistemology after Gettier s famous paper continues to grapple with the problems posed by the GCs to the tripartite definition of knowledge as JTB. As discussed earlier, one of the reasons for this is the fact that a significant number of philosophers consider the GCs as successful refutations of the JTB thesis. This has resulted to at least two things: (1) some form of skepticism about the entire project of providing a definition (or an analysis) of knowledge, and (2) the continuous attempts by philosophers to identify a fourth condition to the JTB so as to avoid the problems posed by the GCs. In any case, it would not be implausible to maintain that contemporary epistemology is still unable to free itself from the clutches of the GCs. For instance, regarding (2) above, David Moshman points out that while philosophers offer many proposals for the fourth condition, no consensus has emerged [6, p. 13]. At this point, one might ask if the tasks in (1) and (2) above are the only tasks left for philosophers to deal with after Gettier. This is where I find the overall goal of this paper to be both important and interesting. By applying Hintikka s SE to the GCs, we can reveal an important but neglected approach to epistemology an approach that considers questioning as a knowledge-seeking procedure (or activity). To illustrate the significance of this kind of approach, let us compare it with the kind of approach involved in Gettier s paper. In the GCs, we find ourselves confronted with the task of trying to determine whether or not a particular epistemic agent has knowledge given certain conditions. In other words, in the GCs, our focus has been on the evaluation (or assessment) of knowledge (i.e. whether or not an epistemic agent S, for instance, knows that P). However, what we fail to consider in the GCs is the process by which epistemic agents form their beliefs or the process

John Ian K. Boongaling: Hintikka s Socratic Epistemology 29 by which they arrive at knowledge. To be sure, knowledge is important (or valuable), but the same thing might be said about the process of acquiring it. In short, applying Hintikka s SE to the GCs will allow us to see knowledge acquisition in a different light. That is, the knowledge acquisition process should be seen as a questioning process where epistemic agents deal with various pieces of information, employ different moves, and make different choices (e.g. strategies) in determining for themselves what to believe in or what they can claim to have knowledge of. 2 Gettier s Counterexamples and the Refutation of Knowledge as Justified True Belief In this section, I will discuss the GCs in relation to their alleged refutation of the tripartite definition of knowledge as JTB: (JTB Thesis): A subject S knows that P (where P is a proposition) if and only if the following conditions hold: (1) P is true, (2) S believes that P, and (3) S is justified in believing that P. For a fuller understanding of the JTB thesis above, several points need to be emphazised. First, the conditions included in the formulation above are to be understood as conditions that are individually necessary and jointly sufficient for knowledge. This means that, if the above formulation is correct, then anything that qualifies as knowledge must fulfill the three conditions specified in it. Second, the abovementioned formulation of the JTB thesis is concerned with what it means for a subject S to know a proposition P. What this means is that the JTB thesis above is concerned not with knowledge in general but only with a specific kind of knowledge: propositional knowledge. Noah Lemos defines this kind of knowledge as the knowledge of facts or true propositions [5, p. 2]. Third, and if we can grant that the second point is correct, then the accomplishments of the GCs should be limited to a particular analysis of propositional knowledge. Fourth, in the formulation of the JTB thesis above, it is important to note that we are interested in some sort of relation between a subject S and a proposition P. Lemos succinctly describes such a relation in the following: We may think of belief as a relation between a subject and a proposition. If the proposition one believes is true, then

30 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 one s belief is true and if the proposition one believes is false, then one s belief is false. We may also think of propositional knowledge as a relation between a subject and a proposition. More precisely, propositional knowledge is a relation between a subject and a true proposition. [5, p. 2] At this juncture, I wish to add that Lemos is not alone in viewing knowledge as a relation between a subject and a proposition. Consider what Linda Zagzebski says in the following: Knowledge is a highly valued state in which a person is in cognitive contact with reality. It is, therefore, a relation. On one side of the relation is a conscious subject, and on the other side is a portion of reality to which the knower is directly or indirectly related... Propositions are either true or false, but only true propositions link the knower with reality in the desired manner. So the object of knowledge in the sense of most interest to philosophers is usually taken to be a true proposition... In a state of knowledge the knower is related to a true proposition. The most general way of characterizing the relation between the knower and the proposition known is that she takes it to be true, and this relation is standardly called the state of belief. [9, pp. 92-93 (emphasis added)] The four points mentioned above are important so that we may avoid, as early as possible, some of the possible misunderstandings/misconceptions that may arise from the JTB thesis and the GCs (most especially in relation to their alleged achievements). For the purposes of this paper, they also serve another important function. They serve as guides that may help make the task of elucidating and evaluating the GCs in a more efficient and fruitful manner. In arguing that the JTB thesis is false, in the sense that the three conditions stated therein are not sufficient for knowledge, it is instructive to take note of two important points provided by Gettier before presenting the GCs. He states: I shall begin by noting two points. First, in the sense of justified in which S s being justified in believing P is a necessary condition of S s knowing that P, it is possible for a person to be justified in believing a proposition that is in fact false. Secondly, for any proposition P, if S is justified in believing P, and P entails Q, and S deduces Q from P and accepts Q

John Ian K. Boongaling: Hintikka s Socratic Epistemology 31 as a result of this deduction, then S is justified in believing Q. [1, p. 121] In his paper, Gettier provides two GCs. In GC 1, Gettier presents the case of two individuals, Smith and Jones, applying for a certain job. Gettier then asks us to grant that Smith has strong evidence for the following conjunctive proposition: (SI 1 ): Jones is the man who will get the job, and Jones has ten coins in his pocket. (SI 1 ) above stands for Smith s Inference No. 1 and it is a label that we will attach to the conjunctive proposition in GC 1 so that we can easily discuss and refer to the said proposition later on. Gettier provides us with some of the possible pieces of evidence that Smith has for (SI 1 ): (SE 1 ): The president of the company told Smith that Jones would be selected. (SE 2 ): Smith counted the coins in Jones pocket (ten minutes ago). Take note that (SI 1 ) above is an inference made by Smith based on (SE 1 ) and (SE 2 ). (SI 1 ) entails the following: (SI 2 ): The man who will get the job has ten coins in his pocket. To complete the first part of GC 1, Gettier asks to grant additional assumptions: (1) that Smith sees the entailment from (SI 1 ) to (SI 2 ), and (2) that Smith accepts (SI 2 ) on the grounds of (SI 1 ) for which he has strong evidence. For the first part of GC 1, Gettier concludes that Smith is justified in believing that (SI 2 ) is true. In the second part of GC 1, Gettier introduces the characteristic move involved in the GCs: the introduction of evidence (or data) that is currently unavailable to epistemic agents like Smith. Gettier asks us to consider that unknown to Smith, he himself, not Jones, will get the job. And, also, unknown to Smith, he himself has ten coins in his pocket [1, p. 122]. This move completes the GC and the resulting picture is as follows: (1) (SI 2 ) is true, (2) Smith believes that (SI 2 ) is true, and (3) Smith is justified in believing that (SI 2 ) is true. If the JTB thesis is correct, we should be willing to accept that in GC 1, we have a genuine case of knowledge since the three conditions in the JTB thesis have been satisfied. However, in GC 1, we are not willing to accept that Smith

32 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 knows that (SI 2 ) is true for the following reasons: (1) (SI 2 ) is true in virtue of the number of coins in Smith s pocket, (2) Smith does not know the number of coins in his pocket, and (3) Smith s belief that (SI 2 ) is true is based on counting the number of coins in Jones pocket whom he falsely believes to be the man who will get the job [1, p. 122]. In GC 2, Gettier presents the case of Smith, Jones, and Smith s other friend Brown. Just like in GC 1, Gettier asks us to grant that Smith has strong evidence for the following proposition: (SI 1): Jones owns a Ford. Here are the possible pieces of evidence that Smith has for (SI 1): (SE 1): Jones has always owned a car. (SE 2): Jones car has always been a Ford. (SE 3): Jones offered Smith a ride while driving a Ford. Gettier then asks us to imagine that Smith has another friend, Brown. Smith is ignorant of Brown s whereabouts. Smith selects three-place names at random and constructs the three propositions below: (SI 2): Either Jones owns a Ford, or Brown is in Boston. (SI 3): Either Jones own a Ford, or Brown is in Barcelona. (SI 4): Either Jones own a Ford, or Brown is in Brest-Litovsk. Each of the propositions above is entailed by (SI 1). On the assumption that Smith recognizes such entailment and then proceeds to accept (SI 2), (SI 3), and (SI 4) on the basis of (SI 1), which, as we granted, is a proposition which he has strong evidence, then Smith is therefore completely justified in believing each of these three propositions. Smith, of course, has no idea where Brown is [1, p. 123]. To complete the argument, Gettier asks us to imagine that the following conditions hold: unknown to Smith (1) Jones does not own a Ford (e.g. He is merely driving a rental car.) and (2) Brown is in Barcelona. Just like in GC 1, we find ourselves confronted with the following problem in GC 2 : There is an important sense in which we can maintain that Smith does not know that (SI 3) is true, even though (SI 3) is true, (2) Smith believes that (SI 3) is true, and (3) Smith is justified in believing that (SI 3) is true. Earlier, we stated that some philosophers consider the GCs as successful refutations of the JTB thesis. Given the clarifications that I made at the beginning of this section, I hope that I was able to provide the appropriate context for interpreting such claims. There is something

John Ian K. Boongaling: Hintikka s Socratic Epistemology 33 however that we cannot really deny: the GCs have changed the landscape of contemporary epistemology. As is well known, in the years following the publication of Gettier s famous paper, we find ourselves dealing with different versions of the GCs. In my estimation, these different versions are themselves proofs for the kind of impact that Gettier s short paper made on contemporary philosophers, especially those in the analytic tradition. In addition, the GCs have become very influential in the sense that they are generally being regarded as decisive or near-decisive tests [3, p. 23] for theories of knowledge post-gettier. Before I end this section, I would like to mention something that might have significant pedagogical import. As I have stated earlier, the GCs have been very influential to the extent that many philosophers created their own versions of it. Thanks to Zagzebski, a recipe for constructing the GCs have been made readily available for all of us: [S]tart with a case of justified (or warranted) false belief. Make the element of justification (warrant) strong enough for knowledge, but make the belief false. The falsity of the belief will not be due to any systematically describable element in the situation, for if it were, such a feature could be used in the analysis of the components of knowledge other than the true belief, and then truth would be entailed by the other components of knowledge, contrary to the hypothesis. The falsity of the belief is therefore due to some element of luck. Now emend the case by adding another element of luck, only this time an element which makes the belief true after all. The second element must be independent of the element of warrant so that the degree of warrant is unchanged. The situation might be described as one element of luck counteracting another. We now have a case in which the belief is justified (warranted) in a sense strong enough for knowledge, the belief is true, but it is not knowledge. [10, p. 69] In the foregoing passage, it is important to note that Zagzebski did not only provide us with a recipe for constructing the GCs. At the same time, she also explicitly mentions an important element involved in the various versions of the GCs: the element of luck. For Zagzebski, this element is crucial in diagnosing what went wrong in the GCs: What generates the problem for JTB, then, is that an accident of bad luck is cancelled out by an accident of good luck. The right goal is reached, but only by chance [10, p. 66]. In the following section, I will now discuss Hintikka s SE with the overall aim of applying it to the GCs.

34 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 3 Hintikka s Socratic Epistemology Explained In section 1, I described Hintikka s SE as revolutionary for the main reason that it envisions a way of doing epistemology without the concepts of belief and knowledge. Before I discuss the rudiments of Hintikka s SE, allow me to expound first on the revolutionary aspect of Hintikka s SE. Philosophers, epistemologists in particular, have always been interested in defining concepts like knowledge and belief. In the case of knowledge, the goal, we might say, is to be able to identify the conditions that are individually necessary and jointly sufficient for such a state. This goal, however, is theoretically misguided according to Hintikka. In addition, he considers the attempt to define the concept of knowledge in a general epistemological theory as an exercise in futility [2, p. 30]. He further adds that such a goal is not useful in practice and useless for applications [2, p. 30]. Allow me to explain the reasons behind Hintikka s seemingly harsh remarks on the attempts to define the concept of knowledge in a general epistemological theory. First of all, Hintikka maintains that the concept of knowledge belongs to applied epistemology [2, p. 30] and not to general epistemology. In my estimation, the reason for this is as follows. In a general epistemological theory, we are concerned with (1) belief -formation and (2) knowledge acquisition. For example, in a general epistemological theory, we might be interested in questions as to how, in normal (or ordinary) circumstances, the data provided (or furnished) to us by our senses usually lead us to form beliefs that are properly warranted. In an important sense, what we are concerned with in any general epistemological theory is the process by which we arrive at beliefs and knowledge. The crucial point that I wish to accentuate here is that beliefs and knowledge are arrived at, that they are themselves results of certain processes. In Hintikka s case, they are results of the process of (interrogative) inquiry. As Hintikka maintains: if knowledge is going to be the end product of interrogative inquiry, it cannot be one of the means of reaching this goal [2, p. 9]. For Hintikka, the concept of belief suffers the same fate as the concept of knowledge [2, p. 30]. To further clarify the points raised above, consider the following example. Suppose John is given a number of propositions by his colleagues about what transpired in a certain conference, which he, unfortunately, missed. Suppose that these propositions constitute an inconsistent set (i.e. the propositions cannot all be true). Suppose that John is determining for himself what to believe in (given the apparently conflicting data in his possession). How will John benefit from a definition of knowledge in this case? In the same vein, how can the concept of belief be

John Ian K. Boongaling: Hintikka s Socratic Epistemology 35 helpful here, when, in the first place, what John is trying to determine is precisely what to believe in? At this point, we might ask ourselves if the concepts of knowledge and belief are indeed unhelpful in this scenario, if these concepts are really indispensable in a general epistemological theory. As I have mentioned earlier, the crucial point to consider is that beliefs and knowledge are arrived at. In other words, they are themselves results of certain processes. In John s situation, we may say that the moment for believing (or knowing) does not yet arise (for him) since he is still in the process of inquiring. This shows that we should distinguish between the results of inquiry and the process of inquiry itself since the latter is the means by which we arrive at the former. This also shows that the role of the concept of belief and the concept of knowledge only come into the picture after the process of inquiry has been carried out. For example, if John is ready to act on a belief that he formed after the process of inquiry, then this is the point where it would matter a great deal whether or not his belief, for instance, is true (or if it is properly warranted). In my estimation, scenarios like this one, can help provide the needed context as to why Hintikka considers the concepts of belief and knowledge as not useful in practice and why they are dispensable in a general epistemological theory. Consider what he says in the following regarding the dispensability of the concept of belief in a general epistemological theory: If you are inspired by this line of thought to review the structure of the interrogative inquiry with a view to finding a role for the notion of belief there, you will not find such a role. Receiving an answer and incorporating it into one s interrogative inquiry is not the same as adopting a new belief. Acceptance is not the same thing as belief-formation... For one thing, at no stage of an interrogative inquiry are there any indications whether or not the inquirer is prepared to act on the truth of the propositions that the inquirer has at that stage accepted (and not bracketed). Hence the entire theory of knowledge acquisition can and must be developed without using the notion of belief. This notion does not play any role in an interrogative inquiry, only in the evaluation of its putative end-point. If one thinks about it, the notion of belief does not play much of a role in the methodology of science. What I am suggesting is that it should not play any more of a role in general epistemology either. [2, pp. 30-31]

36 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 Indeed, one might say that these are very strong words but as we shall see, Hintikka s tough stance against defining the concepts of knowledge and belief (in the traditional sense) and giving them roles in a general epistemological theory is properly motivated. First of all, for Hintikka, knowing is an achievement verb [2, p. 32]. Like knowing, he maintains that believing is also an achievement notion [2, p. 32]. These insights led him to believe that knowledge and beliefs are themselves products of inquiry [2, p. 33]. If this is correct, i.e. if knowledge and belief are themselves products of inquiry, one might therefore ask to why these concepts should play a role in the very method that helps us to secure them. I must admit that viewed in its proper context, and perhaps, with an open mind, Hintikka s insight is a powerful one. Indeed, and as the foregoing discussion shows, SE is revolutionary and philosophers should at least rethink their overall research strategies in epistemology [2, p. 35]. Another important motivation for Hintikka s SE is his critique of Platonic methodology. In the following passage, he summarizes the fundamental assumption of such a methodology and its similarity with SE: I suspect that it is only in Plato s dialogues that he was looking for a definition of knowledge. And Plato put this question (and other questions of definition) into Socrates mouth because Plato shared the widespread assumption that the definition of X gives us the blueprint that enables us to bring about X... Hence it is seen that Plato had in one important respect the same focus as we: the quest for knowledge rather than the justification of beliefs. The definition of knowledge was thought of by Plato as a means for this quest. If so, the pursuit of the definition of knowledge would indeed have been the alpha and omega of epistemology. [2, p. 35] As Hintikka states in the passage above, the similarity between his approach and Plato s is a very limited one, i.e. in terms of what they seek to achieve, both Plato and Hintikka s SE focus on the quest for knowledge rather than the justification of beliefs. It is also worth noting that in the passage above, it is plausible to maintain that Hintikka makes a distinction between Socrates and Plato s Socrates (in the dialogues). This is confirmed by the fact that Hintikka considers Socrates as a practitioner of SE [2, p. 35] whereas he rejects the fundamental assumptions of Plato s methodology. This is confirmed by what Hintikka says in the following:

John Ian K. Boongaling: Hintikka s Socratic Epistemology 37 But we do not think in that way... For us, the fact that knowledge can be considered the end product of inquiry shows on the contrary that it cannot play any role in the process of inquiry. Hence the wild goose chase of the definition of knowledge only shows that too many contemporary epistemologists are still bewitched by Plato s assumptions. This is one of the reasons why... I called contemporary academic epistemology antiquated. Maybe it is time for its practitioners to take up some more up-to-date problems. [2, p. 35 (emphasis added)] After showing the revolutionary aspect of Hintikka s SE, some of the motivations behind it, and its possible repercussions to contemporary epistemology, we will now turn to the task of identifying the elements of Hintikka s SE. In the process, we will also be able to lay bare SE s structure. It is important to note that Hintikka conceives of SE as a game [2, p. 7]. Games, of course, have players. This is the first important element of SE. In SE, Nature and Inquirer 2 are involved in a game (of inquiry). Essentially, we might say that the game consists in drawing inferences and asking questions. The second important element of SE (conceived as a game) is its aim: In SE, the inquirer seeks to determine the truth value of certain propositions in a given model. In ordinary contexts, this model represents the actual world. This means that in SE, we find the inquirer aiming to establish, for instance, a preset formula, say C, from an initial set of premise/s, T. From a methodological standpoint, this means eliminating all the other cases where C obtains (or if we prefer, eliminating all the other C-worlds (or C-scenarios)). The third important element of SE are the moves that the inquirer makes in the course of the game: deductive moves and interrogative moves [2, p. 222]. In the following, Hintikka provides an initial description of SE s elements similar to the current discussion: The basic features of a game of inquiry are clear. A play of the game begins with a number of initial premises. Whenever the presupposition of a question has been established, the inquirer may ask it. If an answer is forthcoming, it is added to the list of available premises. (This presupposes, of course, an analysis of the question-answer relation and a specification as to which answers are available in the particular game in question.) Such interrogative moves can be interspersed by logical inference moves. [2, p. 222]

38 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 One may of course, compare SE to the task of constructing proofs in various areas of mathematics and logic. The main difference however lies in the fact that unlike the task of constructing proofs in the aforementioned areas, in SE, it is legitimate to make interrogative moves aside from purely deductive ones. One may therefore say that in SE, we find not only an original method, but more importantly, a method that better captures what actual inquirers do when they engage in the process of inquiry. At this point, the elements of SE are not yet complete. In particular, we still need to provide an important distinction that Hintikka makes in relation to the rules of the game. This is the distinction between the definitory rules and the strategic rules of the game. For example, in the game of chess, some of the definitory rules of the game pertain to rules related to the following: (1) what counts as a legitimate move in the game (e.g. the King can only move one square in any direction, the King cannot move himself to a square where he can be checked (or captured)), and (2) what counts as winning or losing in the game (e.g. when a player s King is put into check and cannot get out of check, then that player loses the game). As the name itself suggests, definitory rules may be said to define the game. There is something distinct however with regard to how these rules define a game: they tell us which moves are permitted in the game. Definitory rules are thus, according to Hintikka, merely permissive [2, p. 45]. To highlight the difference between definitory rules and strategic rules, consider once more, the game of chess. The (definitory) rules of the game define the game of chess. The same rules, however, do not by themselves tell a player anything about what he or she (or it, if the player is a computer) should do in order to play well, to increase one s chances of reaching the goal. Such advice is what the strategic rules of a game provide to a player [2, p. 7]. Questioning, in the context of seeking knowledge, is not a simple process but at least we can agree that the kinds of questions that we ask greatly affect the kinds of answers that we get. For instance, in the process of questioning, it is not only important for us to be able to ask questions that we are permitted (or allowed) to ask. It is also important for us to be able to ask questions that can best help us achieve (or reach) our epistemic goals. Another important element of Hintikka s SE that sits well with what actual people do in actual problem solving scenarios has something to do with certain rules (or norms) regarding questions and answers: bracketing (and unbracketing). Consider what Hintikka says in the following:

John Ian K. Boongaling: Hintikka s Socratic Epistemology 39 In the typical application of interrogative inquiry for instance in the cross-examination of a witness in a court of law the inquirer cannot simply accept all answers at their face value. They can be false. Hence we must have rules allowing the rejection, or as I will call it, the bracketing of an answer, and rules governing such bracketing. [2, p. 223] In Hintikka s SE, if an answer seems dubious (to the inquirer), then it can be disregarded or bracketed. Of course, all the other answers that rely on the bracketed one must also be bracketed together with their logical consequences [2, p. 3]. The good thing about SE and the element of bracketing is that such a move (i.e. bracketing an answer) can be done temporarily. This means that the bracketed answer can be reinstated (or unbracketed) as a result of further inquiry. SE then can be thought of as a self-corrective process [2, p. 3]. At this point, one may observe that there is a wealth of substantial insights that can be drawn from the questioning procedure in Hintikka s SE with the proviso, of course, that we take the questioning procedure seriously. To complete the elements of Hintikka s SE, we have to delve deeper into what Hintikka refers to as the logic of questions and answers [2, p. 115]. Here, Hintikka introduces several other elements involved in SE: (1) presupposition, (2) desideratum of the question, and (3) the conclusiveness condition of the answer [2, pp. 115-116]. To highlight these elements, Hintikka provides an example [2, pp. 115-116]. Suppose person A asks the following question: (1) Who will win the gubernatorial race? What is the intended epistemic result of asking (1) above? If we follow Hintikka, the intended epistemic result of asking (1) above is a scenario where person A will be able to say the following truly: (2) I know who will win the gubernatorial elections. The specification of the intended epistemic result in (2) above is what Hintikka refers to as the desideratum of the question. It is what is being required from any answer that one may provide to questions like (1). Being able to identify the desideratum of a question is extremely important and it reveals to us something important as well to the logic of questions and answers: It determines the logical behavior of the question in question as well as the behavior of its answers [2, p. 115]. Suppose that in response to (1) above, person B says the following:

40 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 (3) The democratic candidate. According to Hintikka, given several other assumptions, (e.g. that B is honest, that B is knowledgeable about such things), then person A can say the following truly: (4) I know that the democratic candidate will win the gubernatorial race. Hintikka notes however that (4) above is not yet what person A wanted to accomplish in asking (1). It is also not the desideratum nor does it imply the desideratum. Hintikka correctly notes that the reason for this is because person A may fail to know who the democratic candidate is. For the answer to be conclusive, Hintikka maintains that the following should obtain: (5) I know who the democratic candidate is. Speaking of the content of (5) above, Hintikka says: This requirement is called the conclusiveness condition of the answer. It is the presupposition of an answer mentioned earlier [2, p. 116]. At this point, allow me to clarify something that might cause a possible misunderstanding. Why is it that we still see the word know in (2), (4), and (5) above if, according to our previous discussion, Hintikka sees no role in it in general epistemology? In other words, why is the knowledge operator still present in (2), (4), and (5)? Answering this question requires a little bit of context. As I have said in the earlier parts of this paper, SE is a method of inquiry that Hintikka developed through the years. In the previous years, Hintikka refers to it as the Interrogative Model (of inquiry). It is in his 2007 book, the main material used for this paper, that he officially calls it SE. In general, the aforementioned book is a compilation of Hintikka s works on the interrogative model (or method) of inquiry. Some of the chapters (or materials) included in that book have been published previously, and some of the chapters are new. Consistent with what we discussed in the earlier parts of this section, Hintikka maintains that the concept of knowledge does not belong to a general epistemological theory. For clarity, and so as not to miss important details, it is better to let Hintikka speak for himself: The epistemic operator needed in the logic of questions and answers is therefore not a knowledge operator in the usual sense of the term. My emphasis on this point is a penance, for I now realize that my statements in the past might have

John Ian K. Boongaling: Hintikka s Socratic Epistemology 41 conveyed to my readers a different impression. What is involved in the semantics of questions and answers is the logic of information, not the logic of knowledge. This role of the notion of information in interrogative inquiry is indeed crucial, but it does not involve epistemologists usual concept of knowledge at all. [2, p. 26] After clearing up a possible source of misunderstanding regarding the presence of the knowledge operator in the examples above, let us move on to questions and their presuppositions. In SE, the questions that the inquirer may ask are the following: (1) propositional questions and (2) wh-questions (i.e. questions involving what, who, where). Let us begin with the presupposition of propositional questions. Suppose our inquirer is asking the following: 3 (6) Is it the case that S 1 or is it the case that S 2? According to Hintikka, the presupposition of (6) is the following: (7) (S 1 S 2 ). In SE, we have to take note that we can only appropriately ask (6) in the event that (7) has been established otherwise we run the risk of not being able to ask the question in a meaningful way (i.e. helpful in the process of inquiry). This is also true in the case of wh-questions. Consider the following: (8) Who murdered Smith? In (8), the presupposition is clear, i.e. that someone murdered Smith. As stated earlier, we can only appropriately ask (8) in the event that its presupposition has been established. If we will not uphold such a rule (or norm) in the process of inquiry, what then can we hope to achieve by asking who murdered Smith if he has not been murdered to begin with? (Here, of course, we can imagine possible scenarios where Smith ends up dead but was not murdered.) In general, wh-questions, according to Hintikka, have the following presupposition: (9) ( x)s[x]. Before ending this section, let us point out a possible application of Hintikka s views on the logic of questions and answers. We are all familiar with the following textbook example of a complex question (or the fallacy of many questions):

42 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 (10) When did you stop beating your wife? Here are the important points to consider. First, what is the desideratum of (10)? The desideratum of (10), following Hintikka, is a specification of the time by which the respondent in (10) stopped beating his wife. Having identified the desideratum of (10), this also provides us with a range of acceptable answers to (10), i.e. answers that satisfy the conclusiveness condition. Second, the presupposition of (10) is that at some point in time, the respondent began beating his wife. Clearly, the questioner here must first establish the presupposition of (10) before he can appropriately ask (10) since if at no point in time did the respondent beat his wife, then he cannot really answer the question without getting himself into trouble. Third, questions like (10) are not fallacious in themselves since there are contexts/scenarios where asking (10) is appropriate, e.g. when the respondent is someone who in fact beats his wife. Fourth, and this is related to the third point, questions like (10) should not be assessed on their own, or in isolation. They must be assessed as parts of an entire process of inquiry. As the foregoing discussion shows, SE may also have significant implications on how we view the traditional fallacies (in informal logic and argumentation). In the next section, I will now apply Hintikka s SE to the GCs and see if applying such a method can yield important results to the problems in contemporary epistemology after Gettier s famous counterexamples. 4 Applying Hintikka s Socratic Epistemology to Gettier s Counterexamples In the foregoing discussions, we have seen the motivations, elements, and possible applications, as well as implications of Hintikka s SE to contemporary epistemology. In this section, I will make full use of the tools and concepts in SE s arsenal so that we may see for ourselves how such a method of inquiry may impact the GCs. Since the GCs have the same structure, in this section, I will only discuss GC 1. Again, here are the important elements for our discussion of GC 1 : (SE 1 ): The president of the company told Smith that Jones would be selected. (SE 2 ): Smith counted the coins in Jones pocket (ten minutes ago). (SI 1 ): Jones is the man who will get the job, and Jones has ten coins in his pocket.

John Ian K. Boongaling: Hintikka s Socratic Epistemology 43 (SI 2 ): The man who will get the job has ten coins in his pocket. Take note that (SI 1 ) above is an inference made by Smith based on the evidence in his possession: (SE 1 ) and (SE 2 ). Gettier asks us to grant that Smith has strong evidence for (SI 1 ). Now, if Smith is justified in believing (SI 1 ), and (SI 1 ) entails (SI 2 ), and Smith deduces (SI 2 ) from (SI 1 ) and accepts (SI 2 ) as a result of this deduction, then Smith is justified in believing (SI 2 ). Now that the important elements are in place, let us start applying Hintikka s SE to GC 1, but first, let me point out a minor problem at the outset. Ever since I read Gettier s paper many years ago, I have always wondered why he still needed to mention that Smith has strong evidence for (SI 1 ). If this is really that important in setting up the GCs to refute the JTB thesis, I think Gettier should not be surprised if, in the final analysis, it turns out that Smith does not really know that (SI 1 ) is true. After all, having strong evidence for (SI 1 ) does not guarantee that (SI 1 ) is true or that subject S knows that (SI 1 ) is true. This is because the strength of one s evidence for a particular proposition is a matter of degree, and the same thing might also be said about (SI 2 ). Let us now move on to the task at hand. Consistent with SE s methodology and commitments, I propose the following: (1) that we view GC 1 as a game, (2) that in such a game, we will simply view (SE 1 ), (SE 2 ), (SI 1 ), and (SI 2 ) above as pieces of information within the range of our inquirer s (i.e. Smith s) attention, and (3) that the inquirer s task can be partitioned (or compartmentalized). To do (3), I will make use of frames, (F 1 ), (F 2 ), and so on. Before I proceed, allow me to briefly explain the main idea behind the employment of partitions or frames for the current purposes of this paper. On the whole, it can be said that they serve a very practical purpose: Frames or partitions are used to refer to the various parts (or stages) of the process of inquiry. This is in keeping with the idea that I mentioned earlier, i.e. that questions should not be assessed on their own or in isolation, but rather, as parts of an entire process of inquiry. Let us start with (F 1 ). In (F 1 ), Smith has, at his disposal, the following initial set of information: (SE 1 ) and (SE 2 ). Smith can do a deductive move or an interrogative move. Which move will he choose? Can Smith immediately deduce (SI 1 ) (the conclusion) from (SE 1 ) and (SE 2 )? Strictly speaking, Smith cannot do this absent several additional information such as the following: 4 (1) Everything that the company president says is true 5, and (2) Jones and Smith cannot both be hired

44 KRITERION Journal of Philosophy, 2017, 31(3): 25 56 by the company. 6 Two things might be said about Smith s case in (F 1 ): (1) if we are going to be strict about what Smith may infer based on his set of evidence, then Smith cannot correctly infer (SI 1 ) given that several of the assumptions that we mentioned above are absent, and (2) the problems that we encounter in this case are all related to (SE 1 ) rather than (SE 2 ). But since Gettier asks us to grant that Smith has strong evidence (whatever that means) that (SI 1 ) is true, would Smith now be in a position to know that (SI 2 ) is true? This appears to be the case since: (1) (SI 1 ) entails (SI 2 ), (2) Smith deduces (SI 2 ) from (SI 1 ), and (3) Smith accepts (SI 2 ) as a result of the said deduction. Consider (F 2 ). In this frame, let us grant Gettier s request. In other words, that Smith has already established (SI 1 ). A purely deductive move on Smith s part then would entail (SI 2 ). It is, after all, a simple application of the existential generalization rule. In general, we can say that from the fact that Fa, it follows that ( x)f x. But a purely deductive move on Smith s part is precisely the kind of move that will generate the dilemma for the JTB thesis. This is made possible by the introduction of evidence (or data) that is currently unavailable to Smith: (UD 1 ): Smith himself, not Jones, will get the job. 7 (UD 2 ): Smith himself has ten coins in his pocket. Clearly, and as Gettier hoped, (SI 1 ), (SI 2 ), (UD 1 ), and (UD 2 ) will generate the dilemma for the JTB thesis. Here is where the proposal of partitioning Smith s tasks will prove to be helpful in getting things in perspective. It is important to note that (F 1 ) is a distinct task (or game) from (F 2 ). If we will choose to consider (F 1 ) and (F 2 ) as a single game, then is it not only reasonable that we should add (UD 1 ) and (UD 2 ) to the original set of information in Smith s possession? Doing so however will produce (at least) two complications for Gettier s aim: (1) (UD 1 ) conflicts with (SE 1 ) and if (UD 1 ) conflicts with (SE 1 ), it will affect (SI 1 ), which in turn, will affect (SI 2 ), and (2) as a corollary of (1), we will not be able to bring together the needed ingredients for the dilemma that Gettier directs at the JTB thesis. At this point, another important result needs to be mentioned. Earlier, I said that in order to generate the dilemma for the JTB thesis, Gettier needs (SI 1 ), (SI 2 ), (UD 1 ), and (UD 2 ). Interestingly, he has no need for (SE 1 ) and (SE 2 )! Consider (F 3 ). In this frame, we will simply adopt the important elements and assumptions in (F 1 ) and (F 2 ). The difference is that in this partition, Smith makes interrogative moves aside from purely deductive