Might Alston's brand of deflationism about the (unity and usefulness of the) concept of justification be part of the reason that I read epistemologists these days talking less about 'justification' than about the (excessively?) broad 'positive epistemic status'?

Hey Jay,

I'll take a crack at this, but I just woke up and started my first cup of coffee.  You and I overhear a conversation in which someone asserts p.  I ask you whether he was in any position to say that as you know him better.  I think the position is pretty straightforward.  Either the guy shouldn't have said what he did or it was acceptable for him to say what he did.  You might say that his assertion was warranted because he had good reason to think what he said was true.  Now, Tim comes along and says that the guy's assertion wasn't warranted because what he said was false.  But, you say, he had good reasons to think what he said was true.  I have a hard time thinking that you two are talking past one another here.  There's just this question (was he in any position to assert p?) and any attempt to say "Well, in one sense he ought not have said that but in this other sense it is not the case that he ought not say that" will naturally lead to the question "Okay, but what ought he really to have done, spoken or remained silent?" 

If you and Tim aren't talking past one another here, I think it's pretty easy to show that the internalists and externalists aren't talking past one another in arguing about justification.  Suppose you have sufficient justification to believe p.  Could you then be under some purely epistemic obligation not to say that p is true?  I think not, but then I think we have a way of testing proposals concerning justification. 

Maybe the warranted assertion-justification link isn't what I take it to be, but once you see how this is supposed to get the sides to disagree, I think you can appreciate how hard it will be to maintain that these sides are failing to disagree about something.  Forget about assertion and think about the question whether it was wrong for you to have formed the opinion that p was true.  It seems strange to say "Well, your belief was justified, but really you shouldn't have rushed to judgment because ..."  Fill in that with whatever account you like.  Maybe you shouldn't judge because there was evidence you neglected to take account, your judgment was based on information given by an unreliable source, etc...  It seems strange to say "Well, your belief wasn't justified, but there was nothing wrong with believing what you did".  If there's a connection between believing permissibly and justifiably and the question "Is this (really) something I shouldn't believe?" is not ambiguous, then I take it that the internalist gets it wrong if you can satisfy their requirements on justification and still believe something you should not.  The externalist gets it wrong if you can satisfy their requirements and believe something you should not. More likely, the externalist gets it wrong if you can fail to meet the requirements they say are necessary for justification and still it's not the case that you believe something you should not.

Hi Clayton,

Upon starting writing this, I'm not yet sipping my morning tea, myself.  So if I should say something implausible, do take the norms of assertion to be less demanding than what you previously thought. :)

So far I can't agree with you, although I like your case because it forces me to be a bit more concrete and thorough.

I think you're right that the issue of the norms of assertion will probably run parallel to the issues I've raised about justification in the formation of beliefs/opinions--along with, might I add, the 'acceptance' of certain propositions.  So I'll join you in talking (mostly) about norms of assertion.

[

Let's get oriented by starting with your case.  You may want to skip through this illustration, as the points below this illustration are the important part.

Let's give 'the guy' a name: Steve.

What is p?
Let's try: "I will be able to make it to your [Tim's] wedding".  (Later we might also try some assertions in past and present tenses, such as "Tammy [Tim's girlfriend, who was in Steve's graduating class] took eighth grade with me [Steve]" [where it turns out she was gone during 8th grade], and "Germany borders Czechoslovakia".)

Tim is angry at Steve, on September 25, for having said in July that he would be able make it to Tim's wedding on September 24.  Steve was unable to make it because a very important task came up at his work that he alone could take care of.  I'm defending Steve because I know that Steve didn't know the task would come up in July when he made the promise. 

Whether Steve should have made the assertion will depend on a few factors: i) how important it was that his assertion would be true; ii) the likelihood that his assertion would be true, or the weight of the reasons in favor of that assertion; iii) the importance and iv) the (un)likelihood of the factor which made the assertion false. 

(For instance: regarding (i), it is relevant whether Steve's presence at the wedding was important [was he the best man?].  Regarding (iv), did he really know that his work situation would come up, and regarding (iii), was any probable complication likely to have been important [from Steve's July perspective]?)

If it is the case that Steve should not have made the assertion, then presumably he should have looked for defeaters for his belief before asserting it (whether undercutting defeaters, rebutting defeaters, or both).

[If it helps to vivify the example, we might suppose that Steve is one of two ushers who is supposed to be present at the wedding.]

It doesn't seem that it will be easy for Tim and I to agree on whether Steve should have made the promise.  I'll say, look, Steve didn't have any reason to think that his work task would come up, but Tim will say, well, this stuff is important, and Steve needs to actively check and see whether there will be any conflicts before making his promise.

]

I think that the important thing for us to recognize here is that the norms of assertion aren't easy to pick out.  It may be that they are determinate enough that there is a verdict on every case, and it is simply that it is hard for us to know about and agree on what they are.  Or it may be that in some cases they are not determinate.  Either way, it seems that what fixes the norms of assertion for a case (to the extent that they are fixed) depends on some contextual factors. 

 [[a]] In low-stakes situations, as long as someone is aware of a sufficiently strong reason for thinking that p, it seems permissible to assert that p.  (E.g., for me, here and now: 'the Winn-Dixie near my house often offers 2 for 1 sales on fish'.)  This is (at least roughly) the kind of positive epistemic status which internalists call justification.  Now, if it turns out that I actually have a bad memory regarding my Winn-Dixie experiences, and thus am not using a reliable process, it seems that even so I am permissibly making my assertion even without checking my memory of my Winn-Dixie experiences.  That's so, unless I'm talking to someone who rather badly needs a sale on some fish, in which case (if my phone is not working) my becoming justified might require that I go knock on the door of my neighbor, a Winn-Dixie employee who doesn't work in the fish section, and have her quiz me on Winn-Dixie facts to see if whether I'm reliable in my reports on their sales.

[[b]]  In other situations, as long as someone is using a reliable reasoning process when she comes to the belief that p, it seems permissible to assert that p--even if she has not explicitly verified that it is a reliable process.  (E.g., suppose a bank loan officer is using a statistical prediction rule, as her bank instructed her, to determine whether a person is eligible for a loan, and she deems the person eligible.  If the person ends up defaulting on her loan, it does not seem that the bank officer is criticizable for giving out the loan, since she was simply doing her job, and happened to be using a rule which predicts risk better than human judgment does.)  This is (at least roughly) the kind of positive epistemic status which externalists call justification.

[[c]]  In higher-stakes situations, it seems someone needs to look for defeaters (undercutting and/or rebutting) in order to permissibly or non-irresponsibly make her assertion.  (E.g., an army scout in enemy territory needs to make darn sure that the enemy isn't waiting for them in order to suggest marching the troops over the hill.  Even if the scout is using a reliable process to look for enemy troops, the scout also seems responsible for staying alert and attentive for foils to his process, especially if there's the possibility that the enemy might know what his process is.)  In this case, the sorts of positive epistemic status required for justification seem to be a combination of what internalists and externalists call justification.

(I think that the factors which determine stakes would include at least {i} the practical import of the belief (in terms of various norm types, including substantive and procedural rationality, morality, and etiquette) and {ii} the evidence available to the believer that her belief will be subject to scrutiny.)

You write: "It seems strange to say 'Well, your belief was justified, but really you shouldn't have rushed to judgment because ...'."  
--  Yes.  But it doesn't seem strange if you replace 'was justified' there with 'had positive epistemic status'.  In other words, this doesn't seem strange to me: "your belief had some positive epistemic status, but really you shouldn't have rushed to judgment because ...'."  

You write: "It seems strange to say 'Well, your belief wasn't justified, but there was nothing wrong with believing what you did'."
-- But think of the version of my Winn-Dixie case on which my Winn-Dixie memory is unreliable yet the stakes are low.  If 'justified' in your statement refers to externalist justification, your statement doesn't seem all that strange.  I would just insist that the externalist is wrong to think that the positive epistemic status in question which counts as justification is the externalist notion.

You write: "If there's a connection between believing permissibly and justifiably and the question 'Is this (really) something I shouldn't believe?' is not ambiguous, then I take it that the internalist gets it wrong if you can satisfy their requirements on justification and still believe something you should not."
-- I suppose that's what I am saying: questions about whether someone should or shouldn't believe something are ambiguous between asking about *pro tanto* positive epistemic status and asking about *all things considered* permissibility/justification.

Sorry for writing so much; I'm just a little excited about having recently come to these views.

To extend my ideas a bit, I might like to claim that coherentists and foundationalists talk past each other insofar as, depending on the situation, one, or the other, or both of the positive epistemic statuses advocated by each party is necessary for permissible assertion (and belief). 

If I'm simply making an assertion about what I (seem to) see, foundationalist justification seems to be all I need. 

If I'm making an assertion about what helium atoms are like in an informal discussion with my kid, I'm justified given that my asserted belief has a certain degree of coherence with other things I believe.

However (to borrow a case from David Annis), if I am asserting in a doctoral M.D. exam that polio is caused by a virus, it seems that permissible assertion (and belief) requires not only coherence with other things I believe--and not only *explanatory* relations among the coherent set of beliefs in question--but also a proper grounding on various remembered testimonies and perceived experimental results which constitute controlled experiments in which alternative hypotheses were considered and ruled out.  Similarly, in a philosophy class where we're addressing skepticism about the world we perceive, both foundationalist and coherentist justification would be required (and it still might not be enough).

Jay,

I wrote my Master's thesis contra Alston's thesis (though I did not address Swinburne). You can find my thesis via the following link: https://apps.umsl.edu/webapps/weboffice/ETD/query.cfm?id=r5223

-John Fraiser

I thought Swinburne set this up slightly differently:

'The trouble is that there is a rather large range of kinds of actual and possible beliefs about which the intuitions of philosophers and ordinary speakers of the language differ as to whether or not they are 'justified', and the various internalist and externalist accounts give different answers as to whether those beliefs are justified.' (E.J., 2)

So, there is a 'large range of .....',  about which, or in contrast with which, there are a range of ['various'] accounts.

This seems to me to rather initiate the problem of that book as a perceived clubbing together of some responses -- internalist, externalist --as potentially not properly responsive in the first place to some range of differing kinds. (It's as if, where Swinburne said '... the various internalist and externalist accounts give different answers as to whether those beliefs are justified', one could have put 'anyway' or 'just' somewhere or other.)

John,

Thanks.

What I meant when saying that the internalist and externalist might be 'arguing past each other' is that they are both partially right.  It occurs to me upon perusing your thesis that there are more ways to 'argue past each other' than merely to have a merely-verbal dispute with one another. 

Compare with this case: three blind men feel different parts of an elephant; one says 'it's like a wall', the second says 'it's like a snake', and the third says 'it's like a tree'.  These guys are arguing past each other at least insofar as, if they were to get a fuller picture of the truth of the matter, they would see that they get approximate 'proportions' of it correct (as it were).

I think that something similar is going on between externalists and internalists.  (The following statement from your dissertation seems to agree: "Though there are ways in which particular internalist/externalist debates could be compatible, there are also incompatible ones.")

My view is roughly as follows.  [1] In one loose sense of justification, which applies in some contexts but not in others, either externalist or internalist justification is sufficient for justification tout court.  [2] In other contexts, externalist but not internalist justification is sufficient, and [3] in yet other contexts, internalist but not externalist justification is sufficient for justification tout court.  Finally, [4] in still other contexts, internalist and externalist justification are individually necessary and jointly sufficient for justification tout court.

If I'm right, then the following things are true.  [CASE SD1] Suppose someone says, 'internalist justification is the only legitimate type of justification', and someone else responds, 'externalist justification is the only legitimate type of justification'.  These parties have both asserted false theses.  Their dispute is genuine, but they are both approximately equally right about the nature of justification tout court.  So in that sense, they are 'arguing past each other', even though their dispute is not merely-verbal, but substantive.

[CASE VD1] Suppose, however, that one party says 'internalist justification is necessary for justification tout court' and another says 'externalist justification is necessary for justification tout court'.  These parties obviously don't disagree about anything substantive, and are hence having a merely verbal dispute, since it could turn out that all cases are of type [4].  On my view, of course, this is false--there are also situations of type [2] which the internalist party here is wrong about and, and situations of type [3] which the externalist party here is wrong about.

[CASE VD2] Another case where the dispute would be merely-verbal is this: the externalist says 'justification tout court is thus-and-so', and the internalist says 'no, justificaiton tout court is thus-and-such'--but the externalist is thinking of cases of type [2], while the internalist is thinking of cases of type [3].  Then they're talking in the abstract, using similar terms, but the claims they make are not about the same types of scenarios.  That seems like a merely-verbal dispute.  I think that this happens a lot in the literature (although I can' t claim to have mastered the literature on this).

What is tricky is that, for some cases, like [2] and [3], one party is completely right, and the other completely wrong.  But switch to a case of type [1], and both parties are right that their version is sufficient for justification (but only partially, since theirs isn't the only sufficient kind).  And switch to a case of type [4], and both parties are right that their version is necessary for justification (but only partially, since theirs isn't the only necessary element).

Note that, if what I have said is correct, a lot of the substantiveness of future disputes about internalism and externalism will turn on which cases are of each type, [1]-[4].  But the advantage of this framework is that single examples won't be taken as decisive for settling the debate completely.  Numerous cases must be considered--some where the stakes and likelihood of challenge to the belief are very low, others where they are much higher, and others between these extremes.

So: kumbaya.

What do you think?

All sciences and philosophies are based on some foundational assumptions which are not themselves ever proven to be right, they are just the starting point for the thought process. For Newton it was "space and time are absolute".

Whenever people argue from different foundational assumption, they will always be arguing past each other because they begin with different foundational assumptions. As Kuhn explained, this is a paradigm debate, and paradigm debates cannot be resolved by the normal rules of logic, because they are pre-logical assumptions that seem so self-evident that people can't imagine how anyone could possibly believe otherwise.

The externalist begins with the naive realist assumption: I see the world around me. The world I see is the world itself. Anyone who would contest what is plainly obvious to anyone must simply be mad!

The internalist notices retinal after-images after seeing a camera flash. He sees visual illusions. At night he dreams up whole worlds of fantasy. At some point his mind "clicks" and he realizes that everything is an internal theater. Then another "click" : Beyond the farthest things we perceive is the inner surface of our true physical skull, beyond which is the real world, of which all this here is a miniature internal replica.

The two views cannot both be right, one is definitely more factually right than the other, it makes specific predictions (there are pictures in the brain) while the other invokes magic to explain after-images hanging in space instead of on our retina, and dreams or hallucinations where there is nothing there to be directly perceived.

But in the vein of your discourse, there are circumstances, e.g. driving a car, where it would be inappropriate to think like an internalist and see the world around you as an illusion, so you become a naive realist till you get safely to your destination.

On the other hand in philosophy class it is inappropriate to be a naive realist, because your job is to educate your fellow man of the true epistemological configuration of the world of experience being contained within the skull.

When talking to your grocer about the price of beef, it is totally irrelevant whether you are thinking like an externalist or internalist. So in that sense, the choice is relative to the situation, as you suggest.

Nevertheless, the world of experience *IS* in fact entirely contained within your head, that view is actually right, even if the other is sometimes convenient.

If the blind men were feeling parts of a dismembered elephant, they'd be right, its like a tree, like a snake, like a wall. But if they are all feeling parts of the same intact elephant, they are all wrong, an elephant is like a snake and a tree and a wall, all at the same time, and in a specific configuration.

Naive realism is only right inside the world in your head, where you do indeed experience your sense data directly, unmediated. But it is wrong in the larger context, because that sense data is not the object that you think it is, it is a picture in your brain, a representation, not reality. And there really are pictures in the head.

Hi, it is clear to most that there was a time when creatures sensed only an external world - and it was accepted as such. More recently, people - not just philosophers - have came to realise that there is also a "world" inside our heads. And, that that internal world is a mirror image - or  whatever the viewer/philosopher chooses to call it - of the external. Surely there are two worlds: and it is possible to grasp a knowledge of the external by seeing - or gradually coming to understand - the internal. Although we can arrive at a more secure understanding of the internal, or psychological world, the external must persist. If it doesn't, then what will we be able to sense? More important, what then would be the point of having the senses? Over time, the old ideas usually fade. If a paradigm shift occurs - Kuhn - to the internal, then will it be able to eradicate that external world also?