Wednesday, 12 December 2012

K is for Knowledge and metaphorical chocolate

And we're back on track! Here's K for Knowledge (which was supposed to go up on 11th December, with L for LolKants as today's post, oops). Despite all the confusion I do feel I have adhered to the reality, if not the principle of advent calendars: y'know, opening the wrong door on the wrong day, forgetting about it for a few days and then gorging on tiny little chocolates. So, enjoy the metaphorical chocolate!

This one doesn't require a huge amount of philosophical knowledge, but for those of you who are interested (hi mum and dad!) you may be interested to know the definitions of the terms in question here. A posteriori is another term for empirical, a posteriori knowledge is that which requires experience to justify it, e.g we know from experience that water boils at 100 degrees centigrade, but only after we've tested our hypotheses. A priori knowledge is knowledge which doesn't require external justification, either because it is innate, or already contained in the terms in question . The famous example is the phrase 'all bachelors are unmarried' which is true by definition. Maths can also be said to a priori in the same sense (although that is more controversial).

If what Being is saying is true though, we have a problem, most famously elucidated by David Hume:  it seems that facts can either be a priori, known beforehand and somehow necessary, (what Hume calls 'relations of ideas'), or empirical a posteriori facts known in experience ('matters of fact'). The problem is that there's no a priori reason why empirical facts should be related, or should continue to be related. We might find the next time that we boil water that there is an unknown variable which alters the boiling point. It is not logically necessary that water boils at 100 degrees C.

Kant of course, tried to bridge this fork by claiming that we also have synthetic a priori knowledge, that our perception (which requires external stimulus) is already internally structured and therefore contains a priori categories (I.e, in order to say that we 'perceive' something we have to presuppose a priori structures of space and time and causality etc). But that's a whole other  kettle of very complicated fish. Let's just look at Tim shaking his booty!

Tomorrow, M is for Marcuse!

Becca x


  1. "It is not logically necessary that water boils at 100 degrees C."

    Ah, but the Celsius scale is defined by the temperature at which water boils. Whatever temperature water boils at is 100 C by definition, so it's a priori!

    There has been some good back and forth on the related issue of whether the official kilogram is 1kg a priori or 1kg a posteriori. I think it goes to show that trying to classify all knowledge into one of these two buckets is a misguided effort.

  2. haha! Yes, good point. And the debate about whether or not maths in analytic or synthetic is another good example of that issue I think, but I dont know huge amounts about it....

  3. What I find most interesting about this exchange is that even though Carl has dismantled your still works in conveying the point. I've got it in my mind no matter what the temperature of boiling water actually "is."*

    Anything that picks at the notion of being able to know anything is fun...or funny if it's a cartoon.

    *212 degrees of course.

  4. I've been thinking about this and i don't think carl's example works without undermining itself.

    'Whatever temperature water boils at is 100 degrees by definition'

    Well, you could believe that, but I think you'd be believing something that wouldn't be scientific belief as we know it. Under scientific conditions this definition would be inevitably incompatible with the readings on the thermometer.(''It says 95 degrees on the thermometer but what it means is 100 because the water is exhibiting all the signs of boiling'') At least occasionally (eg. if we went to Denver to do this test) you'd have to choose between your own pre-decided idea, and the external verification shown on the thermometer. You could chose to go with your pre-decided idea but I don't see how that is compatible with the methodology which establishes Celsius in the first place. It seems it would be a different sort of statement to a scientific one.

    I think this statement relies on an invalid move from a singular existential statement: 'This water boils at 100 degrees on this scale' to a universal statement: 'All water necessarily boils at 100 degrees celsius' (or 'Hey guys lets set up a system called Celsius which is defined by the fact that water boils at 100 degrees!'). But the problem remains that we haven't boiled all the water in every possible place in the world so we cannot state that the boiling point of water is necessarily 100 degrees, or that there is a system where water always boils at 100 degrees by definition. Not without ignoring all the thermometers.

    So the question seems to be a problem of verification (which is of course a major issue). But I don't think you can say that the statement about the boiling point of water can be scientifically said to be a priori. Can it?

    Incidentally I can't really see how a kilogram could ever be convincingly said to be a posteriori. Surely the definition of a kilogram is something that weighs 1000g. If it weighs 999 grams it's just not a kilogram, it doesn't require us to go around the world weighing all kilograms does it? That doesn't even make sense. Am I missing something? (It could be synthetic apriori though, relying on sense data...?)

    Totally willing to be persuaded though. Didn't really mean to go off on one, but sort of did.

    Just don't want you cats thinking that Being & Tim are relativists. We're totally not ;-)

  5. A couple of different things are getting run together here:

    1. The air pressure at Denver is less than the standard 1 atmosphere, so it doesn't count as a counterexample. I left out the full definition in the interest of typing, but 100 C is defined as the temperature at which water boils *at sea level*, etc.

    2. My kilogram example was about the official kilogram weight in Paris, not 1000g = 1kg. The "international prototype kilogram" weighs exactly 1kg by definition now, even though when it was made it was only contingently thought to weigh close to the old kilogram.

    3. When we say "Water boils at 100 C" we can mean (at least) two different things by that. One is "the temperature at which water boils is called '100 C'" which is a definition, and so a priori. The other is "the temperature at which water boils is the temperature called 100 on the C scale 212 on the F scale, etc." The second interpretation is an a posteriori fact, since it is about the temperature itself and not what the temperature is called.

    4. There is an additional complication to the water boiling example. The current definition of the Celsius scale assumes that water always boils at the same temperature given the same air pressure. But suppose it turned out that water boils at a different temperature when light shines on it than not, or on Mondays but not Thursdays. In those cases, the definition of C would no longer work since there would be no such thing as "the temperature water boils at." So, Celsius is a definition (a priori) that requires certain physical truths (a posteriori) to be meaningful.

    5. A priori/a posteriori is not the same as analytic/synthetic or necessary/contingent. (See Wikipedia for more examples on this.) A pri/post is generally used to refer to our mode of epistemic access to a truth. For example, if the laws of physics are necessary (as some people think they are), then it would turn out to be necessary truth that water boils at the temperature we now call 100C, but our only way of knowing this would be through experience, so it would be necessary but a posteriori. On the other hand, analytic/synthetic is about whether we learn any new truths by working through the argument. (If I already knew the meaning of 12, do I learn something new in hearing that 5 + 7 = 12?) Necessity itself is also a slippery concept. Generally speaking, necessity is used to talk about the ontology of the world, not our epistemic access to it. Some people take it that "necessity" just means "true at all times." However, if this is the case, then it's (probably) "necessary" that "a green unicorn is not riding a pink bicycle across Wyoming," since in all likelihood that will never happen. Another theory of necessity is that necessity refers to what is allowed by the laws of physics (so called, "nomological necessity" as opposed to "temporal necessity"). In addition there is "logical necessity" for things that have to be true by the rules of logic, and probably some other forms of necessity I'm forgetting. Long story short, there are many ways to talk about what is or isn't the case and why and how we know it. As such, I don't think it's as helpful to divide things into a priori and a posteriori as philosophers have tended to assume.

  6. Hey Carl I have been so busy that I haven't had time to reply. But I will! This is very interesting stuff....still don't think I agree though ;-)