Greg Detre
Monday, February 12, 2001
Lucy Allais, History of Philosophy V
Is there anything more to A�s causing B than B�s
regularly following A?
Hume thinks that the idea of a necessary connection in
nature cannot derive from either the senses or the understanding (reason).
Assess his arguments for this claim.
Is it possible for an empiricist to give an adequate
account of our conception of causation?
events = constructed parcels of our xp, not real
persons as constructed, like objects, in a panpsychist pantheistic world
choice = choice if we think we have a choice
FW as uncaused/undetermined, doesn't this make *problems* for causation?
in order for the counterfacutals to be regular, tehre must be a (causal) algorithm underlying
at the level of the linguistically construced 'I', I have a choice
what about the game of life?
would the programmer be creating causation, or regularity?
FW and the illusion of choice is emergent
Tempting though it may be to try and learn about the nature of (our) psychological phenomena by introspection (turning our attention inwards), Wittgenstein deems this fruitless: one cannot learn about the nature of sensation by paying close attention to the feeling of a headache. Introspection is not a means to a definition.
Wittgenstein attacks the idea that pain is private in the way that a picture in my head is private. One does not �know� that one is in pain � you just are in pain.
His argument that all that we can know about causality is regularity seems ludicrous, and yet has proved extremely resilient.
deterministic world = one where things were determined to take a single path and could not have been any different given complete knowledge of their starting positions
This is what leads to the problem of determinism in a Newtonian, mechanistic world. It would be interesting to see whether he considers the problems that would result for free will if his project to understand the principles and workings of the mind was successful.
Given then that our notion of causation does not come directly from any impression(s) that we have or could have had, where does it come from?
It is difficult to consider Hume�s theory of causation except within the context of his assumptions about mind and belief. If we discard them as being crude and wrong, then we might be able to say more about causation itself.
In this way, causation has a psychological as well as physical component.
Indeed, the reason that we place so much faith in reason is that we believe that its workings are necessary, that is to say, the laws underlying our process of thought are ultimately dependable, if we are careful enough. The demonstrative derives its force from our conviction that
Here then, we are faced with empiricism taken to its ultimate conclusion, an omni-sensing being, and yet we have still failed to firmly ground our faith in causation. Hume and all subsequent empiricists must choose whether to discard causation or their empiricism.
To say to an alien that something is �red� is no more expressive than to say to a human that something is �asldkgjasiod�. Unless the alien is acquainted with our perceptual mechanisms and some examples of redness, or vice versa, �red� has no meaning, it is a property
At this point, we seem to be trapped in an infinite regress, which either demonstrates that Hume�s requirement for necessary connection is incoherent, or so unfulfillable as to lead us with certainty to an ontological claim about a-causation in the world.
And consensus does not shape reality, only our agreement of it.
With reason, we have an unhappy triangle. Our choice between the incoherence of necessary connection and certainty that there is no necessary connection in the world seems to rest with reason � yet either possibility seems to undermine reason itself.
find out Hume�s exact phrasing for NC
the question then becomes: what would it take for there to be more to causation than A�s causing B regularly???
an observer able to sense the entire world for all time with human senses would only see regularity
but surely, this is just saying that although causation cannot be derived empirically, that it still could exist???
but if so, how would God know causation if he see it???
CC vs NC = difference between regularity and causality
causation = regularity +
divine???
FW???
causation is derived from my experience as a causal agent
causation is derived from introspection into the workings of reason
causation = phenomenologised Custom
is Hume doubting that there is causation in the world, ontologically??? almost certainly not � only that we have no access to it epistemologically
the problem for me is that I can�t imagine any sort of properties of objects outside a framework of laws that would give rise to NC � then we are depending on the uniformity of nature for causation � so the problem of induction and the problem of causation are a pincer attack on all knowledge of matters of fact � together, they completely undermine our knowledge of the world as it is
FW + counterfactuals
jigsaw pieces, snooker balls � as examples of where we can even imagine NC??? kind of like primary qualities???
Grayling pg 504??? says that Hume denies necessary connection and causal roles to thought
We have to trust in the the necessary connection of our thoughts above all, above reason, and that is where induction gets its force.
what difference does the origin of the notion of causation make??? so what if I�ve shown that our notion of causation might be derived from ratiocination itself???
go back to my �causal agent� thing � regularity would be exposed without an algorithm to decide what to do when the universe has to make a choice based on the decision of an agent
I think clearly we have to get rid of necessariness in reality by admiitting that skeptcisism and the ED are most sensible and anythign else is delusion
would we be able to re-cast the omni-sensing observer hypothesis to please a Kantian about time in the world in itself???
if we take our reason as demonstrating necessary connection between propositions, mental objects/events, then it is not a wholly unreasonable step to wonder whether�
I'm going to end by casting aspersions on rationality, and say that I don�t believe any of this claptrap
what if we do we away with this talk of �events�, and consider only a continuous stream of existence � there can never be regularity, since �we can never step into the same river twice�, that is to say, existence never repeats itself exactly � that is why the notion of resemblance is so crucial. does causation work better for a stream than regularity would???
Is it the term �event� that Hume uses, when he talks about A�s following B???
need we have necessary connection, as opposed to simply predefined algorithmic determined connection???
counterfactual-generating algorithm
quantum mechanics makes a nonsense of causality
necessary a priori???
consider whether my attack on reason survives the simply test that we can apply of: demonstrative knowledge = denial doesn�t imply contradiction
if we disregard deductive reason as the only real form of knowledge, then induction (i.e. degrees of knowledge) is no longer subject to doubt
why didn�t a well-considered empiricism like Hume�s consider how we could know anything a priori???
well, he called them �relations of ideas�, as though associationism could deal with the problem
of course, in a way, connectionism is a very very low-level form of associationism
surely you need a mechanism to assess truth, rather than just hodge the ideas together � and what determines the faintness and liveliness of these ideas, if not a rational mechanism???
does Hume consider the problem of FW that would result if his project to understand the principles and workings of the mind was successful???
are counterfactuals like superpositions??? is that the same thing as saying they are possible worlds or quantum states that got collapsed???
where then does causation come from??? simply the innate processing mechanism that he terms Custom???
is there any argument that our reliance upon causation could be based on anything else, e.g. the efficacy of will???
the FW thing helps us because we can say that there must be causal laws to sort out what would have happened if we�d acted differently
without causal laws, i.e. a set of algorithms that �matter� must obey, how could the world deal with counterfactuals???
would that be any more than a finite state machine???
well, if we are free to do absolutely anything we like, it would have to be an infinite state machine.
ah, then couldn�t the world be an infinite state machine, sort of like the Eternal Recurrence, with every possibility already mapped out like chess moves with all the ramified futures completely determined
even if it�s not a big long list but a set of rules, would those then be causal laws though, or just rules defining the regularity???
surely they�d just be rules of the regularity, sort of like painting by numbers � you have an algorithmic sort of blueprint telling you what colours to put where, but unless you are able to perceive the whole picture as a visual image, each individual colour will seem wholly arbitrary. in order for it to be necessary conjunction, there has to be something about the world that means that there is something about the essence of every object that it must have a certain outcome�
can I think of such an example that is an algorithmic regularity law???
surely the point I am making is that if you accept the causal schema of natural laws into which each object is going to fit, then those laws of nature themselves define what must necessarily be so
is this what mackie means by necessity1 and necessity2, i.e. necessary within a causal schema vs necessary a priori???
no
But would God + self-defining laws of nature satisfy us - isn�t there something about the way Hume phrases his definition of necessary connection that implies something even more primary about the objects than simply that the laws operating on them would determine the effect??? like what � what could give the properties of objects necessary power to do something other than the ultimate laws under which they operate???
is causation a coherent concept in a time-continuous world??? how do we separate it up into separate C+E events??? what is an event???
does it make sense to think of Hume�s attack on causation as a sort of epiphenomenalism???
by this I mean that matter is causally inert; it appears to play a causal role, but is in fact just echoing the pattern underneath � no, that�s the opposite of epiphenomenalism
would Descartes� proof of God be synthetic a priori???
what are the physical + psychological components of causation???
what difference does the origin of the notion of
causation make??? so what if I�ve shown that our notion of causation might
be derived from ratiocination itself???
anomalousness of the mental = a problem � does ratiocination ever show necessary connection � certainly we make mistakes all the time�???
Can he not have kind of subconsciously abstracted from the structured, sequential, necessary nature of ratiocination, in the same way that Custom works on experience without our rationally bidding it. Is it any different to talk of Custom as applied internally to reason, as opposed to its application to our experience???
can the primary/secondary qualities distinction help us with regard to achieving a meaningful and helpful conception of necessary connection???