Back    All discussions

2010-02-08
Berkeley and the Passivity of Ideas
It's well known that Malebranche, Berkeley, and Hume among others argued that we don't observe strict or efficient causality; rather we just observe correlations (A and B) and confuse strong expectations of B following on the occasion of A with strict causation of B by A. Yet there is a good deal of psychological literature developing Albert Michotte's experiments (Perception of Causality, 1963 English translation) by those like Brian Scholl (Yale Lab on Cognition) and others that there is a robust representation of causality when observing simulations of collisions. Robust means even in cases when there is no regularity in the sequences, (no correlation) people including infants represent in perception that A moves B in that particular case when a launching of one object by another does appear to take place. Michotte admitted that his work might not have convinced Hume, if representing causality in perception required representing that A was necessarily connected to B in a strong sense comparable to an analytic connection.
As J. Bennett noted about Berkeley,(Learning from Six Philosophers) the claim that ideas are passive might be a priori --that is, it wasn't clear what Berkeley was asking us to look for. With Malebranche it seems clearer; we don't note agency in collisions.(A chair doesn't suddenly decide to move across the room) Berkeley discusses something like this in De Motu. Matter is by definition inert;that is, it doesn't act. But the reproduction of Michotte's experiments suggests humans at least, in a fundamental and quite immediate sense, distinguish causation from mere correlation. (See Phil Quarterly Vo. 59 No 236 July 2009--Symposium on the Admissible Contents of Perception)

My particular interest is Berkeley's notion that ideas (objects of sense) are passive. But any discussion of the this stuff would be appreciated.

Dick Brook
Professor of Philosophy
Bloomsburg University
Bloomsburg PA 17815 

richardbb@ptd.net

2010-02-09
Berkeley and the Passivity of Ideas
Reply to Richard Brook
RE: "Michotte admitted that his work might not have convinced Hume, if representing causality in perception required representing that A was necessarily connected to B in a strong sense comparable to an analytic connection."

This seems to me to highlight the distinction between science and philosophy - so often blurred these days.  One can "observe" all kinds of things, but there is still the philosophical question to be asked: "Through what series of concepts can we provide a coherent, transparent explanation of the event/experience?" Assuming Hume's analysis remains intact (and leaving Kant aside) no amount of observation will allow us to circumvent that question, will it?

This reminds me of the difficulty I have with many contemporary discussions of consciousness (for example). One "observes" all kinds of things (human reactions to various stimuli, monkeys looking into mirrors, people in comas etc etc) but the philosophical problem remains untouched - how to give an adequate conceptual account of consciousness. 

Of course, one solution would simply be to abandon philosophy for science... 

DA

2010-02-10
Berkeley and the Passivity of Ideas
Reply to Richard Brook
Dear Richard,

It seems to me plain that sometimes we are acquainted by direct experience with the causal connection between events. So, when I hear a car backfire
and I jump, I know from that single case that I jumped because of hearing that loud sound. It isn't as though I note the hearing and the jumping
and think maybe there is a causal connection. I know one is there. The reason I know is that, in addition to the hearing and the jump,
I experience the hearing of the backfire causing the jump. In short, causation runs through nature, my mind/body is part of nature
and so I am acquainted with causation running through the part of nature I can introspectively observe. I believe E. Anscombe
noted the startle case.

When I stub my toe and scream I experience the pain making me scream.

Similarly imagine losing an arm wrestling match with a stronger opponent. One experiences one arms being forced down, despite one's efforts at
resistance, but that's experiencing causation. Or getting pushed or shoved. There is more than the sensation of contact and one's
body moving, there is the pushing causing the body to move, the body being made to move by the pushing, one feels that.
Unlike the historical empiricists, I think we live in a world where we constantly experience causation, all of it.
What goes on inside goes on outside too. So there is no problem for an empiricist like Hume
to find an idea of causation richer than constant conjunction. We have impressions of events causing
other events, not merely of conjunctions of events. We often experience causings.

I believe Anscombe gave examples of watching a knife cut cheese--we see the causal process of cheese being separated by the
movement of the knife. Watching a nail being hammered one sees the nail being driven deeper by the blows.

Best, Jim

2010-02-10
Berkeley and the Passivity of Ideas
Reply to Jim Stone
But surely this does not answer Hume's objection?  Observation in itself can never solve the problem he raises, nor any amount of introspection.

DA

2010-02-10
Berkeley and the Passivity of Ideas
Reply to Derek Allan
True; we can't observe necessary connections in nature if by n.c. we mean something like an analytic connection. If A making B happen requires that, then the game is up.Michotte recognized that about Hume.[Is this your point about Hume's objection?]  But that notion of necessary connection is too strong. If we allow natural necessity--why can't we represent in perception that the brick hitting the window made it break. Moreover, Hume's (and Malebranche's and Berkeley's) view of the etiology of beliefs that we perceive causal connections--we observe B following A a number of times and  then strongly expect B to follow A---is, if the results of current experiments developing Michotte's work are correct, mistaken. Early moderns make assumptions about human psychology which are open to empirical challenge. 
Best regards,
Dick

2010-02-10
Berkeley and the Passivity of Ideas
Reply to Derek Allan
True; we can't observe necessary connections in nature if by n.c. we mean something like an analytic connection. If A making B happen requires that, then the game is up.Michotte recognized that about Hume.[Is this your point about Hume's objection?]  But that notion of necessary connection is too strong. If we allow natural necessity--why can't we represent in perception that the brick hitting the window made it break. Moreover, Hume's (and Malebranche's and Berkeley's) view of the etiology of beliefs that we perceive causal connections--we observe B following A a number of times and  then strongly expect B to follow A---is, if the results of current experiments developing Michotte's work are correct, mistaken. Early moderns make assumptions about human psychology which are open to empirical challenge. 
Best regards,
Dick

2010-02-11
Berkeley and the Passivity of Ideas
Reply to Richard Brook
I argued above that we often experience causings in precisely the way David Hume maintained we don’t–I gave the startle reaction as an example.

Let me ladd something still more tendentious. It seems to me that we directly experience causation in sense perception. That is, when I feel a rough surface, sandpaper say, I experience the feeling of roughness as caused by something external to me, the thing that I feel. The experience comes to me as caused by contact with a rough surface. The same goes for most auditory experiences and visual experiences. The visual experience of seeing an automobile bearing down on me comes to me as caused by an approaching object.

This isn’t because of any beliefs I have. From the first, experiences were of causings. It's how sense perception works.
This explains simply why we all believe there is an external world and also why it is a condition of satisfaction of sense experience that sense experience is caused by its object. Just as the experience is a representation of a rough surface, it is a representation of a causing by a rough surface.

Also this helps explain the difference between impressions and ideas. Impressions typically come to us as caused, mere imaginings typically do not. Part of what the empiricists are inchoately gesturing at when they talk about the ‘vivacity’ of impressions is that sense experiences, as opposed to imaginings, are typically experienced as caused. ‘Impressions’ literally come to us, and are experienced as, impressions, that is, effects of external things.

If so, the experience of causation is ubiquitous. 

2010-02-11
Berkeley and the Passivity of Ideas
Reply to Jim Stone
Thanks for your reply Jim and your examples. The examples I am interested in concern mechanical interactions--billiard balls and such. The kind Malebranche, Berkeley, Hume and  others use to illustrate the thesis that matter is inert, that is it doesn't do anything. If, so the claim goes, we carefully observe collisions there is no property--causality--that we observe. In fact our simply causal judgments are projections of strong expectations that B will follow A after we observe B invariably following A. The work of Albert Michotte and those developing it e.g. Brian Scholl-Yale lab on cognition) appear to show this is false. Transitive causal verbs like "move" reflect a robust representation of causality in perception.
There is the question whether any of this empirical stuff cuts any ice against Hume. His argument seems to be: (1) causal relations are necessary connections. (2) We can't observe necessary connections in nature. Therefore we can't represent causal relations in perception.

My own tentative view is that (1) is true, but (2) is false. With Berkeley,my particular interest is, accepting for the sake of argument his view that no object of sense (idea) can exist unsensed, does it follow that idea-idea causation is impossible--say in collision phenomena.

Regards,
Dick Brook

2010-02-12
Berkeley and the Passivity of Ideas
Reply to Richard Brook
Hi Richard

I really need to check the precise terms of Hume's argument which I am a bit rusty about (I am out of my comfort zone on your topic).  But I thought I would ask another question first.

At the risk - probably the grave risk - of distorting your argument, I wondered if in fact you are trying to dispense with the notion of causation altogether? I mean, if one can argue on the basis of Michotte's experiments than subjects often infer causation when there is none (eg the big ball "chasing" the small ball example) perhaps you are suggesting that all perception of causation is questionable and that the very idea misleads us?

Apologies if this is way off target.

DA

2010-02-15
Berkeley and the Passivity of Ideas
Reply to Richard Brook
Perhaps relevant is a distinction many philosophers make (including Hume and Kant) between what we 'strictly' observe and what we observe in a looser sense.

In the looser sense, any reliable non-inferential response to objects of the right kind is counted by many philosophers as a (concept-laden) sense perception of that object.  So, for example, physicists can observe electrons passing in a cloud chamber in this sense.

However, while granting that all perception is concept-laden (perhaps 'theory'-laden) in this sense, one can distinguish within this framework between what one 'sees the object as' (viz., in relation to the concept mobilized in the perceptual response to the object), and what one 'sees of' the object that one sees (for example, one sees the facing surfaces but not the inner pages or back cover of a given book).  It seems to me that this phenomenological/conceptual distinction is required in an account of perception in general, and that it does not commit one to sense-data, etc.

So consider a given magnet.  I 'see of' it its 'proper and common sensible' properties, as Aristotle called them: a given size, shape, color etc.   I have various corresponding sensations of various facing color surfaces, etc. (again, one need not reify these 'surfaces' themselves into 'sense data').   Do I similarly have a visual sensation of the magnet's magnetic properties?   No.  Hume and Kant rightly follow Plato and Aristotle on this point.  I do not see of the object its magnetic properties in this strict sense.  Of course, I immediately see it AS a magnet; I see objects in my world AS having all kinds of dispositional properties, and I observe such properties 'immediately', that is, non-inferentially, as directly causally evoked by the object (given my conceptual 'set').  But I do not see of the object its causal properties in the same sense in which I see the proper and common sensible properties of objects.

Can't this distinction be used to generate Hume's sceptical question in clear ways, e.g. without reliance on either sense-data or on any objectionable assumptions pertaining to the 'passivity' of sense experience, etc.?  (This doesn't address Jim Stone's particular point, although I personally think it could be generalized to cover it.) 

2010-02-20
Berkeley and the Passivity of Ideas
Reply to Richard Brook
Richard is interested in things like billiard ball collisions, which was the sort of case Hume began with, and uses as an example in his Abstract and in EHU. But he generalised his account of our acquaintance with causation as merely experience of regular sequence to the psychological cases, such as Jim's case of being startled by a noise, saying "The uniting principles among our internal perceptions[such as hearing a noise then being startled] is as unintelligible as that among external objects , and is not know to us any other way than by experience." (T 169) Yet when in his Appendix he takes this claim seriously, and sees all his own perceptions as having mere correlations uniting them, not really efficient causation, he feels something has gone wrong, and he has no adequate account of his own mind, and what unites his own perceptions. His initial analysis of the causal relation had found it to consist in temporal sequence, spatial contiguity (where applicable) and necessity, so he may have gone wrong at the start, as its unclear that we do regard causes as necessitating their effects. So he was launched on his quest to track down the origin of the idea of causal necessity, and failed ot check if the idea of "efficacy" really was the same as that. As Anscombe pointed out, we do think we know that the knife pressed onto the cheese cuts it, the first time as much as any subsequent time. Hume did allow that sometimes we are sure of a causal connection after one case, so generalize from that, but thought this was because of our familiarity of hundreds of similar but not quite the same causes.But I think his main error was in taking necessity to be part of what we take causation to be. He may also be wrong in thinking our knowledge of how our own nature works is the same as our knowledge of how the rest of nature works, so he treats his own perceptions as if they were as separate as billiard balls.it possible that Locke was right that our idea of power does originate in our own exercised powers, not just of our wills but our muscles, in pushing, pressing etc. When Hume argues against Locke, it is the will as efficacious cause he dismisses. Of course for Berkely too the will's moves are not passive ideas, and we have notions of them, not ideas. Hume is so set against the Locke-Berkeley priviliging of the human will that he fails to ask if other aspects of our psycholigy might not reveal efficacious causation to us, without displaying necessity. This is just a note on Hume's approach, and does not advance your main concerns,  Annette Baier.    

2010-02-22
Berkeley and the Passivity of Ideas
Reply to Richard Brook
Dick,

Interesting post.  Can you say a bit more about  the "robust representation of causality" in the experiments.  What is the evidence of this?  From what I remember (from years ago), these types of studies were done on small children, and the evidence that they were vulnerable to the illusion of causation was based upon indications of surprise.  Have researchers like Scholl gone farther with this? 

Mark Collier
University of Minnesota, Morris

2010-02-23
Berkeley and the Passivity of Ideas
Thanks for this helpful post.

I have always read Hume, when he says we don’t experience necessary connection, to be saying this: we don’t experience causings or efficacy or makings. We don’t experience the causal nexus itself. I did not read him as saying that we do experience efficacy or something much like it, but it doesn’t rise to some level of necessity that he has in mind. ‘Necessary connection’ meant, I thought, a cause’s making the effect happen, as often occurs. That is what we don’t experience, according to Hume, not even something roughly in the ballpark. We experience constant conjunction and contiguity, nothing extra. The sense that we experience the causal relation itself, or something like it, is an artifact of expectation.

2010-03-02
Berkeley and the Passivity of Ideas
Reply to Derek Allan
Am I correct in thinking that in the period we are considering, Newton's kind of science (I offer no hypotheses - i.e. I do not attempt to explain) was only about observation and that 'explanation' was left to metaphysics? Newton withdrew his hypothesis that gravity was a force acting at a distance on the grounds that it was a step beyond the observations and therefore not good science. Hume, I suggest, would have had no time for 'explanations' as he aimed to apply Newtonian science to human understanding. He did not try to 'hypothesize' about the 'hidden levers and springs' and, if I recollect correctly, put such questions into the category 'nonsense' - having no meaning for us animals. 
Currently, science seems more obsessed with explanation than observation as the ART of modeling possible explanations hits the headlines and the observations are either ignored or suppressed (as in the case of some climate data).

One could go further and point to the hypotheses leading the science  - scientists being funded (bribed?) to find data to support the hypotheses. Patrons of science in the past may not have been quite so demanding and may have sought 'new knowledge' for its own sake or because it revealed something of God's mysterious ways? Or have i got the wrong end of the stick?

Before we abandon philosophu in favour of science, I think we should examine the credentials of much of today's science. 

2010-03-02
Berkeley and the Passivity of Ideas
Reply to Richard Brook
Yes, Jim, Hume did think we experience efficient causality, but only when we know a constancy of conjunction, never from just one case, if that case is unlike earlier experinced causes. So the first time we cut into a cheese, we cannot be sure our pressing the knife really made the cut in the cheese, or the first time a noise startled us, that our reaction was to the noise. And since many of our own perceptions are unique one shot ones, we cannot know if their apparent effects on later perceptions really are cases of causation. Hence, I think, Hume's doubts in the Appendix, about whether his account of our minds as a series of perceptions with causal relations uniting them was really satisfactory. Its all very well to say Adam could not be sure, on seeing his first billiard ball collision, what caused the second ball to move as it did, but to say he could not know if his own concern to see if the same sort of thing happened again was caused by his observation of the first case seems less plausible. We think we do know what sparked our curiosity, when it is sparked, and do not have to wait to establish constant conjunctions before being sure. Since constancy of conjunction is what Hume thinks we confuse with necessity, it was his putting necessity into the causal relation, when he first defined it,at T 77, that led to his later final defintions, at T 170, which require constancy of conjunction, so seem to rule out our knowledge of causation the first time we encounter a certain sort of cause. That is why I think his search for the origin of our idea of necessity led him off the track, if it was causation he really wanted to define. His second definition of cause, as anything perception of which makes us expect a certain follow up, the effect, seems satisfied in the startle case, and the curiosity case, well before we know that the first definition, requirning constancy of conjunction, is satisfied. (We will however, always expect a constancy, where we think we have obseved a cause.) But do the robust experiments show our recognition of causes working on or in us, or our own causings, to be more readily known than causation in the world outside us?  Annette Baier.

2010-03-02
Berkeley and the Passivity of Ideas
Experience of constant conjunction couldn't give experience of causality if experience of causality is experience of necessary connection, and the latter is meant to be like an analytic connection; that we could read off the effect simply from knowing the cause. Moreover if Michotte and his followers are correct we represent examples of causality in perception, for examples B1 moves B2, without, and even against, the constant conjunction of (B1 collides with B2) and (B2 moves). (B=billiard balls) That is adults, children, some animals distinguish in perception transeunt activity from constant conjunction.
Dick Brook

2010-03-02
Berkeley and the Passivity of Ideas
Reply to Collier Mark
Actually they have been done by Michotte and others with adults and replicated in children. A recent issue of Phil Quarterly devoted to philosophy of perception refers to Michotte's work and developments by those such as Brian Scholl at the Yale lab for cognition.
Dick