Wednesday, 7 March 2012

What's so great about evidence?

I’ve just read a paper, Why Do We Value Knowledge? by Ward E.Jones, that poses the question: if we are concerned with truth, why bother with knowledge?  If knowledge is factive then all knowledge is true, but not all truths are known.  Why value a p which is known more than a p’ which is merely true?  Why do we, and it’s pretty evident that we do, think less of the lucky guesser as opposed to the knower?  Both have truth, both believe the proposition in question, the lucky guesser just lacks warrant, or justification, or reasons, or evidence for the proposition.  What’s so great about evidence?

Jones examines the instrumentalist answer to the question; that we value evidence (or warrant, or justification, or reasons...) because it is a means of getting true beliefs. As Jones points out that is a cracking explanation of why we like evidence but no explanation of why we like true beliefs brought about by evidence above those that we just happen upon.  An example from Jones is getting across a river.  We want to be on the other side and nice, safe, reliable method can be depended upon to get us there.  So we will use one. But the state of being on the other side is no sweeter because we got there by a safe and reliable method. 

My own way of putting it is this: we want money, in part, because it lets us buy beer.  But free beer is not valued any less.

Jones’s suggestion is to look at the incidental attributes of knowledge.  In order to get a robust definition of knowledge the concept has been pared down as far as possible, anything non-essential removed.  According to Jones, it may well be in this cloud of non-essential attributes of knowledge that its real value is found.

I think that there is a simpler and more straight-forward explanation.  We value neither truth nor knowledge.  Of course we will say that we value both, but our actions do not bear our statements out.  Rather like my saying that I am fond of a low fat, alcohol-free diet with plenty of exercise is not borne out by my sitting in a pub eating crisps and drinking beer.

Consider four propositions; e1, e2, g3 and g4. The “e” propositions have as much evidence (or warrant...) as you like whilst the “g” propositions are complete guesses. The odd numbered propositions, e1 and g3, are true.  The even numbered propositions, e2 and g4, are complete rubbish.  Our attitude to the propositions will be divided by evidence rather than truth.  We will think e1 and e2 are true, be inclined to believe them and consider them examples of knowledge.  Our attitude will not change dependent upon whether they are knowledge or whether they are true or false.  Similarly we will be wary of both g3 and g4, further investigate both the true and the false, deny both of them the title “knowledge” and be reluctant to believe them to be true.  Whilst we are avowedly all for truth and knowledge, dispositionally our attitudes track evidence.

We do not have access to the truth of propositions beyond the evidence we have for them. And so, the evidence we have for propositions stands as a proxy for truth, it is something we can aim our beliefs at and adjust our attitudes on the basis of. 

Now on the traditional view of knowledge, knowledge is evidenced, or warranted, or justified or reasoned true belief.  Where we have that evidence (or warrant...) we also believe the proposition to be true and, so, believe that all the ingredients of knowledge are there.  Where we do not have that evidence (or warrant...) we do not call our position in relation to the proposition “knowledge”.  Our attitude to propositions tracks our use of the term “knowledge” and, so, we believe that we favour knowledge when, in fact, we favour evidence.


Jones, W. E. (1997). Why Do We Value Knowledge? American Philosophical Quarterly , 34 (4), 423-439.