Tagged: Standard Analysis of Knowledge

The Gettier Problem and Patient Knowledge

As I have now studied the Gettier problem for some time (3 years) and will not continue to go on with it, let me resume my experiences in three short statements.

  1. I did not make the experience that analytic philosophers (at least here in Austria) liked to discuss the Gettier problem with me. This may be due to the fact that the Gettier problem has been discussed within Analytic philosophy for already more than 50 years, and they are fed up with it. It may also have to do something with the fact that the Gettier problem is not exactly a glory for Analytic Philosophy, but rather the opposite. Anyway, it did not seem to me that analytic philosophers were pleased that somebody with a different philosophical orientation showed interest in one of their theoretical problems.
  2. Being an analytic philosopher seems to be very much a question of intuitions; and if one does not share their basic intuitions it is not possible to join their group. For example, I do not share their intuition that in Gettier problems the truth condition is fulfilled. (The truth condition is fulfilled only if the analysis of Gettier cases is narrowed down to the literal meaning of propositions, but not if the intended meaning of the persons who utter those sentences is taken into account.) However, there was no way for me to discuss that issue with analytic philosophers, they just were not interested in it. This phenomenon may be problematical, for if it is necessary to share the intuitions of analytic philosophers in order to enter a career in academia, then becoming a philosopher is not so much a question of rationality but of joining a group of quasi-religious believers (the group sharing analytic intuitions).
  3. The analytic philosophers I have discussed the Gettier problem with did not accept any other method of analysis of it than their own. Their own method consists in inserting the propositions generated by the specific Gettier case into the equation of the so called Standard Analysis of Knowledge. As I, with my background in Phenomenology, came up with my phenomenological method I did not encounter any understanding of its relevance for the Gettier problem from the side of analytic philosophers.


In my paper „What we can learn from the Gettier problem about knowledge. And what follows from that concerning the possibility of patient knowledge” I have tried to show that the emphasis on the aspect of truth within the concept of knowledge leads to a decrease of self-confidence in weaker epistemic subjects and, consequentially, has a disempowering effect on them.

Hofbauer Kaelin Ankersmit Feigl Is the Patient a Human Being 2015

[„Was wir aus dem Gettier-Problem über den Begriff des Wissens lernen können. Und was daraus für die Möglichkeit von PatientInnenwissen folgt“, in: Helmut Hofbauer, Lukas Kaelin, Hendrik Jan Ankersmit, Walter Feigl: Ist der Patient ein Mensch? LIT Verlag, Münster 2015, p. 131-165.]

The argumentation is divided into two parts: in the first part the situation of the individual subject of knowledge is considered, in the second part the social dimension is added.

The Standard Analysis of Knowledge and the Gettier problem sketch the problem of knowledge on the level of an individual subject of knowledge knowing one specific proposition (one sentence). On this level I focus on the fact that all Gettier cases omit the question who supplies us with the information about the truth. We are, for example, told that a person sees a sheep in a field and forms the belief that there is a sheep in the field (the proposition in question), but that this sheep is in reality a dog that looks like a sheep. However, behind a little hill in the same field there actually is a real sheep that makes the proposition “There is a sheep in the field.” true.

In such a case I, as a phenomenological philosopher, would now be interested in knowing the person who brings the news about the real sheep behind the hill, and I would like to ask her under which circumstances she saw it and if she made sure that it really was a sheep. However, analytic philosophers do not have any problem with the concept of a truth that is not noticed and communicated by anybody. This is why in Gettier cases the truth is just told by the auctorial storyteller, and we have to accept that there is a real sheep behind the hill although this information somehow pops out from nowhere.

In my paper I have used this “magic” in the Gettier problem (that we do not know the source from which the knowledge about the truth is provided from) in order to question another central intuition of analytic philosophers, namely the intuition that in order to call something “knowledge” the content of what we know has to be true. In my opinion (and following my intuition), if we say that we “know” something we just want to say that we are quite sure about it, but not that we are 100% sure about it. Because otherwise we could not err.

Remark: This stands in blatant contrast to what analytic philosophers believe: “Everyone agrees that knowledge entails true belief, in the sense that if one knows a proposition, p, then one believes p and p is true. (Of course, one might think that one knew a certain proposition which turned out to be false, but in such a case one would thereby discover that one did not really know it after all).” (Duncan Pritchard: “Knowledge”, online: http://www.philosophy.ed.ac.uk/people/full-academic/documents/KnowledgeCIPFINAL.pdf)

Confusing knowledge with absolute knowledge and with the fact that we also know for sure that we know a specific proposition is hubris. In my paper I have visualized this hubris by a God’s eye that is suspended above the hill and able to see the real sheep hiding behind the hill. By believing in the Standard Analysis of Knowledge, that is, in the claim that knowledge is (a) believing a proposition which is (b) true and (c) being justified in believing that proposition, we are kind of playing to be God for the truth condition in my opinion is too strong for knowledge.

Therefore, there is something wrong already in the Standard Analysis of Knowledge, namely the truth condition, and the Gettier problem just shows this mistake by construing a fold into reality which in this case is represented by the hill that separates the spectator from the sheep. In the original Gettier cases this “reality fold” is built by the unpredictability of the future and by great geographic distance.

It can thus be deduced that persons who have the intuition that something they know is really true (because otherwise they would not know it) are people who are very sure of themselves. They are know-alls, people suffering from the mania of always being right. The phenomenological method actually shows that by indicating to the fact that in Gettier cases it would be necessary for them to use God’s eye in order to conceive the truth.

Gettier case-sheep in the field example philohof

In the second part of my paper I have added the social dimension to this basic situation. The argumentation goes as follows: We, human beings, are not gods, but we are experts and lay-persons, stronger epistemic beings (healthy persons, grown-ups, professors) and weaker epistemic beings (ill persons, children, students), and an expert looks like a god in comparison to a lay-person.

The truth condition of the Standard Analysis of Knowledge therefore is a good criterion for knowledge if the goal is the defence of the established social order. The expert, as a person who is busy all day long in her narrow field of specialization, will always be able to prove to the lay-person that the lay-person does not know anything about a specific topic because she does not know as much about it as the expert.

However, if the goal is to empower the weaker epistemic subject, as is case e.g. in patient knowledge, the truth condition is too ambitious: From the expert point of view the ideas of a lay-person and, even more, of a chronically ill person suffering from pain and shortness of breath, must inevitably appear to be irresponsible.

Therefore, in order to make something like patient knowledge possible – that is: to establish knowledge of weaker epistemic subjects – knowledge would have to be defined differently than in the Standard Analysis of Knowledge. The truth condition which is only useful for the search of mistakes committed by weaker epistemic subjects in order to make them ashamed of themselves would have to be substituted by one or more other condition(s). Possible candidates are: whether additional knowledge about their illness enables patients to change their behaviour, whether it adds up to their already existing knowledge, whether it empowers them to get along alone for longer periods without panicking and rushing for expert’s advice, whether it improves their self-confidence, and so on.

In summary, the intuition shared by analytic philosophers (and they claim that this intuition represents common sense) that something we know has to be true in order for us to know it, is only true for over-self-confident people, for know-alls and dogmatics. Patient knowledge construed after the model of this intuition will result in permanent control of patients by doctors via technical devices like cell phones. The reason for that is that is appears irresponsible to let somebody alone who knows less than you do (that is: who is not right all the time, at least according to your own standard).

The situation would be different if we ventured to define knowledge as the ability of a person to get along by herself – and only in then something like patient knowledge would be possible. However, in this case we would have to renounce from our tendency of judging this person every time she commits an error. For the knowledge of an autonomous person is not judged by the truth condition of the Standard Analysis of Knowledge but by her ability to get along by herself.

Discursive Closure

In this post I want to explain my concept of discursive closure. As example I can use the Gettier problem. Every topic  one might choose excludes other topics. That’s inevitable. But it makes a difference, of course, how people are talking about a specific subject. If they are talking about it in a way that produces the Impression that other aspects of the same problem do not even exist or are not relevant all, it is a case of discursive closure. Discursive closure, in my Definition, means that a certain discourse of a certain class of persons draws a borderline around the problem  inhibiting to understand the problem in other contexts or simply to understand it differently. In short, discursive closure closes a problem towards new and other ideas.

Take for instance the Gettier problem: which problems does it exclude? First and foremost it excludes the problem of the quantity of knowledge. That’s clear, because the Gettier problem is about propositional knowledge. It asks what it means to know one single proposition.

But what, if you make the experience that other people deny you knowledge because of the fact that you do not know all about a certain subject? I vividly remember the experience at the beginning of my philosophy studies at the University of Vienna when in a class on Plato’s Phaedrus I heard the statement that Plato should be read in Greek in order to be properly understood. This statement shocked me because it devaluated in advance the knowledge we were about to acquire in this specific class.

It happens quite often that knowledge is devaluated because a person does not possess all or, at least, very much knowledge about a specific discipline or field of research. Actually, we are attributing knowledge to experts because they know a lot about their field of expertise. To be an expert does not mean to know one specific proposition and to be sure about it but to know many things about a certain field of knowledge. And we take the mass of known propositions concerning a certain field of knowledge as warrant for validity of specific propositions and of the epistemic judgements of the expert.

Or take another example: in the method of Structuralism, one should know and make explicit all relations inside a specific field of (e.g. social) relationships in order to determine the character of one specific relation or action of a subject inside this field. This methodological requirement produces the meed for a lot of work in research and privileges the more hardworking scientist and the one with the more comfortable access to the data before the more intelligent one.

If you have problems of that kind and you come to the Gettier problem, then there is just no possible way to thematize them. Imagine that because of this problem you feel the urge to talk about knowledge. When you arrive at the experts of knowledge, the first thing they might do is ask the question “What is knowledge?” and respond to it with the Standard Analysis of Knowledge (Knowledge is justified true belief.). The next step will already the objections to the Standard Analysis of Knowledge brought forward by Edward Gettier’s counterexamples and the many responses to it by other philosophers.

When this happens you are already trapped. You might want to say: “But I want to talk about a different problem concerning knowledge.” And the others might respond: “Talking about such an important problem like that of knowledge, you have to start at the beginning – and the beginning is the Standard Analysis of Knowledge and the Gettier problem.” If you take them seriously you will become old before arriving at discussing YOUR problem.

Of course I know that Analytical Philosophy is something like the ideology of common sense. Like common sense is convinced that the human being is capable to perceive and know one specific thing in the world without all the other things (Quine is an exception here.), Analytical philosophy, too, starts with simple elements in order to study them in more complex combinations later on. That means, this is a certain Approach. Each philosophical school has its specific approach and as such it is ok to begin the study of the concept of knowledge by studying the knowledge of single propositions.

But even though I would kindly ask to keep in mind the concept of discursive closure. It easily might be forgotten that beginning with simple objects is only the approach of a specific philosophical school. It can easily happen that the conviction gains the upper hand “that this just has to be done like that” or that in this field of knowledge just one specific approach is valid.

When this happens many people will not be able to formulate and discuss their problems in philosophy, and also those people will become professional philosophers who have right kind of problems that fit into the framework of allowed or accepted problems of their discipline.

This is why we should remember the concept of discursive closure which assures us that every problem or question – like, for instance, the Gettier problem – is not just there to be solved, but also to cover up and to make invisible other problems and questions.

We can only always talk about one thing at the same time. But that does not mean that there are no other things at the same time.

On the Gettier Problem

At this years Wittgenstein Symposium (11-17 August 2013) I will give a talk on the Gettier problem. I am still thinking hard about what I want to tell the people who will attend my talk because the Gettier problem is a real case of what I call ‘Philosophical Intercultural Communication’. By that I want to say that the congress participants (I expect them to be professionals in philosophy, experts) will think that what I am saying about the Gettier problem is absolute nonsense, whereas, on the other side, I, myself, think that the Gettier problem is absolute nonsense.

So, there seems to be no possible way of communication. And if, as we suppose it to be in philosophy, it can only be the case that I am right and they are wrong or they are right and I am wrong, then it follows from that that either they are or I am an idiot, stupid or little intelligent, which is the component of shame in philosophical discussion.

Thus, the fact that I go there in order to speak about something – the Gettier problem – I do not understand, just in order to express my total incomprehension of it, can by regarded as courageous, silly, or as an act of sheer aggression against the discipline of philosophy.

But it is neither of these attributions. It is really the case that professional philosophers – be it in English or in German – are often talking in a “different language” from mine, and I do not understand what they are saying. And at the same time, while listening to them, I have the Impression that, if their way of talking about the subject is correct, then there is no place left over for my thoughts to be expressed (because their way of conceiving the problem or topic makes my treatment of it impossible from the very start). Or, to express it in other words, it is really a case of Philosophical Intercultural Communication.

The Gettier problem for me to is a parade example of the huge differences between the different cultures of philosophizing, because it seems to be necessary to go through a kind of mental conversion (similar to a religious conversion) for a person in order to be able to understand the Gettier problem at all.

I will now give a very short synopsis of the Gettier problem as I understand it. The aim of the Gettier counterexamples is to prove that the so called Standard Analysis of Knowledge is insufficient. The Standard Analysis defines Knowledge as justified, true belief. A person S knows a Proposition p iff (I) she believes p, (II) if she is justified in believing p, and (III) if p is true. In his article from 1963 Edmund Gettier constructed two counterexamples in which all of these three conditions are fulfilled, but even though the person S does not know.

The first counterexample is about Smith and Jones having a Job interview. Smith is justified to believe that Jones will get the Job. (Why? That’s not so important. It might be “that the president of the company assured him that Jones would in the end be selected.”) And Smith has seen ten coins in Jones pocket. Therefore he draws the following inference: “The man who will get the job has ten coins in his pocket.” But the truth is that Smith, and not Jones, will get the Job, and Smith also has ten coins in his pocket (but he does not know that.

In conclusion, Smith believes that “the man who will get the job has ten coins in his pocket”, he is justified in believing it, and it is true – but, even though, he does not know it.

That’s at least the Interpretation of the case of the analytical epistemologists. They say, Smith was right, but luckily right. There was too much luck involved, epistemic luck, and he could, under different circumstances, easily have been wrong. And the fact that his assertion was nothing more than something like a ‘lucky guess’, is the reason why we say that Smith did not know that  “the man who will get the job has ten coins in his pocket”.

What I do not understand about this Interpretation of Gettier’s counterexample is the following: It is obviously based on the supposition that linguistic utterances have to be taken literally. That means that “the man who will get the job has ten coins in his pocket” can either be true or false, and in the case that it is true, all three conditions of the Standard Analysis of Knowledge are fulfilled.

That’s not my way of thinking. I would ask Smith: “What do you mean by ‘the man who will get the job has ten coins in his pocket’?” As I am neither a logician nor a lawyer, “the man who will get the job has ten coins in his pocket” and “the man who will get the job has ten coins in his pocket” can mean different things, depending, for example, on the man, Smith had in mind when constructing that phrase.

Or, shorter, I would not Interpret the first Gettier counterexample in the way that Smith was right when assuming that “the man who will get the job has ten coins in his pocket”. What he assumed, had the same linguistic expression like what was true, but it had a different content. What Smith had in mind was: “Jones will get the job, and Jones has ten coins in his pocket.” What was true was: Smith got the job, and Smith had ten coins in his pocket.

The reasons why the Gettier problem is important for me is, first, to be confronted with people who really do think logically. They think logically means that they take linguistic statements literally. For them there is no difference between “the man who will get the job has ten coins in his pocket” and “the man who will get the job has ten coins in his pocket”. (If you understand me literally, I will get angry. This is also the reason why I get angry with the Gettier problem: I also get angry when I experience that other persons – even fictitious ones are understood literally. I think that we express our contempt towards a person when we understand her literally and not according to what she wants to say.)

And, secondly, the Gettier problem shows in which kind of problems and mess you put yourself when thinking logically. The Gettier problem is up to now unsolved. And there is a reason for that: the concept of knowledge expresses an inner relation between the knower and the thing she knows. That is: something additional to external conditions like those of the Standard Analysis of Knowledge.

It can be evaluated by an observer whether a certain person has a certain belief (for instance if she says to have this belief), if there are reasons for her to be justified in having that belief, and if the belief is, in fact, true. But in order to find out what it is a person knows one has to start asking her what she means when saying that she has this certain belief. But logically thinking philosophers do not do that, for the logical analysis of linguistic expressions for them is already the highest degree of accuracy possible in philosophical research.

Knowledge has to escape them, for knowledge is an inner relation. Or, the inner relation is an important part of its characteristic: “I know that I know.” “I know what I know.” If you want to find out about what somebody knows you will have to start asking this person about what she thinks to know. If you start – like the epistomologists and logicians – with the conviction that the same words always necessarily mean the same things, and that it is completely irrelevant what people have in mind when saying them, you will miss knowledge. Of course, you will not miss other things which can be seen from outside, like truth, belief or justification – but you will miss knowledge.

This is what is really interesting about the Gettier problem: that logical thinking leads to an absolute standpoint (a perspective in which statements can just be true or false – and not: interpreted differently), and that there are certain things – like knowledge – which cannot be understood from such a standpoint; that knowledge pertains to those queer things which require a second, a relative perspective in order to be understood. Relative perspective means: You have to go, yourself, to the place where a certain person stands in order to find out what the landscape looks like from that standpoint.

In conclusion, the Gettier problem is interesting for me because so many intelligent philosophers writing a mountain of sophisticated papers have failed solving it. But instead of suspecting that there is something wrong with the conception of the problem, they are still investing more logic and sophistication into it.

Now I am about to come and tell them that the king is nude and to take away their favourite toy from them. How will they react? I suppose that by proving to me that I know so little about formal logic, the rules of logical thinking and handling definitions that I will degenerate psychically to a sobbering something that does not even know its own name, anymore.