tag:blogger.com,1999:blog-9192597712746432631.post5147708131281993849..comments2024-03-09T09:06:35.288+00:00Comments on Notes from Two Scientific Psychologists: Chemero (2009) Chapter 7: Affordances, etc (Pt 2)Andrewhttp://www.blogger.com/profile/16732977871048876430noreply@blogger.comBlogger10125tag:blogger.com,1999:blog-9192597712746432631.post-64815954242198390512011-04-17T12:56:11.170+01:002011-04-17T12:56:11.170+01:00Frazier,
I've read some of Bickhard's stuf...Frazier,<br />I've read some of Bickhard's stuff and like where he is coming from. There is some question (in my mind) of whether it does every we want, but it definitely does the "process" thing. I, personally, suspect that a satisfactory answer will do a strong tie back to W. James, and contextualize Eco-psych within a wider context of psychology. Of course, the biggest difficulty is that to do that psychology needs a coherent identity.... ah, problems. <br /><br />EricEric Charleshttps://www.blogger.com/profile/17412168482569793996noreply@blogger.comtag:blogger.com,1999:blog-9192597712746432631.post-66855245279015597622011-04-17T12:08:18.750+01:002011-04-17T12:08:18.750+01:00But if you don’t believe me (as Andrew doesn’t), I...<i>But if you don’t believe me (as Andrew doesn’t), I hope you will agree that we need a view of affordances that is genuinely dynamical. </i><br />I think this is fine; but I still think that the place to do this is in information. What ebbs and flows and changes over time is us, and therefore our perception of the properties of the world. I must blog in detail about task specific devices, because I think that's where the action is.Andrewhttps://www.blogger.com/profile/16732977871048876430noreply@blogger.comtag:blogger.com,1999:blog-9192597712746432631.post-32068634882841485132011-04-17T12:05:52.905+01:002011-04-17T12:05:52.905+01:00Let's look at A with the salt and water exampl...<i>Let's look at A with the salt and water example. When we say that salt has properties that make it dissolvable, we have not fully specified the disposition: Salt has properties that make it dissolvable IN WATER (and other suitably propertied liquids). In the same manner, when we say that the stairs are walk-up-able, we have not fully specified the affordance: The stairs are walk-up-able FOR ORGANISMS OF A CERTAIN SIZE AND SHAPE. If you always list who the affordance is for as part of the affordance, you will have most (if not all) of the essential features of the relational view. A "who it is for" is an essential part of the description of the affordance in both systems.*</i><br />As I said above; actually, you can tell a story about the solubility of salt, even if water doesn't exist, if you have a theory about how the anchoring properties lead to solubility. This is a part of industrial chemistry: you have a material with known properties, and a theory of, say, solubility, and you can build a solute to match. I think this sounds more like Gibson's 'pointing both ways', and it certainly gibes with the affordance implying the organism and vice versa. I like how such an implication is very strong and robust for the dispositional account; relations can do it to, but it seems less...compulsory about what is required to fill the implied complement slot.<br /><br />But I don't want to argue on the basis of quotes, I've come to the conclusion it's a bad idea. Here's what I think is ok about dispositions: they are properties of the world and don't become affordances until measured by an organism (via perception); or, more lately, measured by scientists. I worry about relations as simply not being the right kind of thing to lead to information, and nothing Tony's said convinces me otherwise. While the difference may end up small, I think the difference in starting assumptions (affordances as features vs. properties) matters and I much prefer the property story. I'm going to try and find time to work up a summary post to try and convey this.Andrewhttps://www.blogger.com/profile/16732977871048876430noreply@blogger.comtag:blogger.com,1999:blog-9192597712746432631.post-129662960410442262011-04-15T00:23:38.928+01:002011-04-15T00:23:38.928+01:00"There could be better ways there. Lately, I’..."There could be better ways there. Lately, I’ve become more and more convinced that Bob Shaw is onto something when he talks about the need for a process ontology."<br /><br />...Or Mark Bickhard's call for the same, which has been rather extensively elaborated.P. Adrian Frazierhttps://www.blogger.com/profile/04214839055576596218noreply@blogger.comtag:blogger.com,1999:blog-9192597712746432631.post-55609247482201122682011-04-14T13:29:46.680+01:002011-04-14T13:29:46.680+01:00Andy, Certainly no more thorough reply is owed. It...Andy, Certainly no more thorough reply is owed. It is nice to have a place to think through these things. <br /><br />I was thinking about this last night, and was going to post something similar to what Chemoro just said. What I realized was 1) if you specify the "affordance" thoroughly in the dispositional view, you are really close to the relational view. 2) so my problem with the dispositional view must be A - in encourages (but in no way mandates) an incomplete specification of the affordance, B - it has that ugly effectivity thing, C - the implication of necessity. Point A was the big one.<br /><br />Let's look at A with the salt and water example. When we say that salt has properties that make it dissolvable, we have not fully specified the disposition: Salt has properties that make it dissolvable IN WATER (and other suitably propertied liquids). In the same manner, when we say that the stairs are walk-up-able, we have not fully specified the affordance: The stairs are walk-up-able FOR ORGANISMS OF A CERTAIN SIZE AND SHAPE. If you always list who the affordance is for as part of the affordance, you will have most (if not all) of the essential features of the relational view. A "who it is for" is an essential part of the description of the affordance in both systems.*<br /><br />So, part of my problem with the effectivity-thing is that it is ugly (it lumps together several things that I think it is virtuous to treat seperatly), but, more importantly in this context, it is redundant. If we apply the above principle in reverse, a proper specification of an effectivity would include a description of the behavior those animal properties allow. Thus, if you fully specify either one, you don't need both. <br /><br />The necessity thing may be a red herring, as we could specify sufficently to make the necessity argument make sense (I just think we shouldn't).<br /><br />--About the dynamic thing. I'm not sure what to think about this. Surely we miss things when we think statically, but surely much is there statically. Is it a question of where the best starting point is? Is it a question of where we are destined to end up no matter where we start?<br /><br />--<br /><br />*Worth noting, Holt went through great length to argue the object towards which a behavior was directed was an essential part of the description of a behavior - many interesting implications of this.Eric Charleshttps://www.blogger.com/profile/17412168482569793996noreply@blogger.comtag:blogger.com,1999:blog-9192597712746432631.post-35579454244385447002011-04-14T03:57:19.172+01:002011-04-14T03:57:19.172+01:00Andrew mentioned something in a comment a while ba...Andrew mentioned something in a comment a while back that is relevant here: Mike Turvey and I have been doing some modeling of differing views of affordances, and find that the relational and dispositional views are in fact very similar. The deeper you dig here, the more similar the views get. The real difference comes when you compare Gibsonian and representationalist views of affordances. Donald Norman, Vera & Simon, Ruth Millikan, and Erol Sahin all develop views of affordances that take them to be kinds of mental representations. Affordances on these views are fast mappings of perceptions to actions, which can occur without a lot of intervening computational work. They are, that is, what Andy Clark calls action-oriented representations. When held up against this work, the differences between affordances-as-dispositions and affordances-as-relations become truly insignificant. <br /><br />(This is, I think, in contrast to the differences between the situation-semantic and TSM views of information. There, the differences are consequential.)<br /><br />As regards what I call affordances 2.0 in the book, it really doesn’t matter whether you start from relations or dispositions. I prefer relations, so I started there. I gave some reasons in the book that I prefer to understand affordances as relations, and I have a few others. But if you don’t believe me (as Andrew doesn’t), I hope you will agree that we need a view of affordances that is genuinely dynamical. So, develop a dynamical view of affordances based on a dispositional view of affordances, and what will it look like? Well, I actually worked it out a while back and can tell you that it will look a lot like the version of affordances 2.0 based on relations. (I can’t find the graphic I made of it at the moment. I will post it if I do.)<br /><br />This is the main point, as far as I’m concerned: Given what ecological psychologists do, we can’t define affordances in terms of things that are static, as relations and dispositions are. Affordances 2.0 is an attempt to get there. There could be better ways there. Lately, I’ve become more and more convinced that Bob Shaw is onto something when he talks about the need for a process ontology.Tony Chemerohttp://edisk.fandm.edu/tony.chemeronoreply@blogger.comtag:blogger.com,1999:blog-9192597712746432631.post-33227537764530141552011-04-13T20:04:57.189+01:002011-04-13T20:04:57.189+01:00I'll come and go on these points as I have tim...I'll come and go on these points as I have time and in no particular order; sorry, long week and I've been thinking about this in pieces :)<br /><br /><i>Your analysis here does not really tout the virtues of the dispositional approach, it merely states that the relational approach is not necessary for much of what Chemero wants.</i><br />This is true. I am currently working on the premise that the dispositional view is the orthodoxy here and Chemero's is the challenger. So I've been focused on addressing the problems in the factors motivating his account. So a more positive story is certainly due: I've done bits of one <a href="http://psychsciencenotes.blogspot.com/2010/05/affordances-part-3-dispositions-or.html" rel="nofollow">here</a>, and <a href="http://psychsciencenotes.blogspot.com/2011/02/fcking-affordances-how-do-they-work.html" rel="nofollow">here</a>. I owe you more, though :) For now, within the context of reading Tony's book, I'm focused on using the dispositional account as a basis from which to address his concerns.<br /><br /><i>In the TSM model, talk of affordances is just (i.e., only) a shorthand for talking about clusters of object properties. If objects exist, and have properties, then they have TSM-style affordances, QED. The fact that there is sustained metaphysical interest in the notion of affordances indicates there is something in the notion more interesting than the TSM version. The relational version emphasizes that more interesting part. </i><br />Yes. I think this is in Gibson too: the central question, after all, is not 'are there affordances?' (because yes, there are), but 'is there information for them?'. As I've mentioned, information is the relational bit, and I think that indeed, that's where a lot of the action is. The problems Tony poses that requires a relational solution (e.g. the problem of two minds) gets handled in optics, not in the world. <br /><br />I'm reading the Cutting paper: I think he thinks Gibson is up to relations. But I think that the 'affordances point both ways' bit fits nicely with the dispositional framework. Think of salt: the anchoring properties of it's disposition to dissolve implies quite specifically the required properties of the solvent. This type of information is what you use to make new solvents, or new glues, etc. There's a clear specification of the complement in a disposition that relations lack; lots of things can be the relata of any given relation. This needs some work, but I think there's meat to these bones.<br /><br /><i>As for the “uninvestigatable” assertion: There ARE things that one can do within the TSM paradigm, but the one thing you definitely CANNOT do is investigate whether or not an-organism-with-the-effectivity takes advantage of the-affordance. In the presence of an affordance, an organism that does not act upon it, by definition, lacks the effectivity. In the presence of an organism with the effectivitiy, if the action does not occur, by definition, the affordance is not present. Those two, true by definition, statements are the primary intended meaning of the disposition analogy / metaphor.</i><br />This strikes me as entirely unproblematic. If an organism effects an affordance, then the presence of the affordance and the capacity to effect that affordance are, indeed, trivially the case. The interesting bit has always been the information: without that, you can't clearly lay out the full nature of the effectivity. What you study, then, is not 'whether' an organism does something, but 'how'. Sure sounds like the ecological research programme to me.Andrewhttps://www.blogger.com/profile/16732977871048876430noreply@blogger.comtag:blogger.com,1999:blog-9192597712746432631.post-15651913953742758932011-04-11T03:15:21.618+01:002011-04-11T03:15:21.618+01:00Well,
Your analysis here does not really tout the...Well, <br />Your analysis here does not really tout the virtues of the dispositional approach, it merely states that the relational approach is not necessary for much of what Chemero wants. I have already talked in other posts about problems I have with the dispositional approach. First, I think the notion of effectivities is highly problematic. Second, I think the notion of necessity in realizing the “disposition” is highly problematic. Third (and final for this list), I think at best the disposition formula is true by definition, and hence, at some level uninvestigatable. <br />Another point, which is NOT fair as a criticism, but is useful for comparison, is that in the TSM model, the metaphysical reality of affordances is uninteresting. In the TSM model, talk of affordances is just (i.e., only) a shorthand for talking about clusters of object properties. If objects exist, and have properties, then they have TSM-style affordances, QED. The fact that there is sustained metaphysical interest in the notion of affordances indicates there is something in the notion more interesting than the TSM version. The relational version emphasizes that more interesting part. In the relational view, the affordances IS the matching between object properties and organism abilities. The affordance is that I can walk through the door – not something about the door, not something about me, but something about the match between us. This, among other things, sticks us to a physical interactionism, making it clear that our primary concern is my physical abilities, and not my knowledge or predilections (which get lumped together in effectivity-talk). It also takes out any notion of necessary action; a relation is there, but there is no disposition that MUST realize. <br />As for the “uninvestigatable” assertion: There ARE things that one can do within the TSM paradigm, but the one thing you definitely CANNOT do is investigate whether or not an-organism-with-the-effectivity takes advantage of the-affordance. In the presence of an affordance, an organism that does not act upon it, by definition, lacks the effectivity. In the presence of an organism with the effectivitiy, if the action does not occur, by definition, the affordance is not present. Those two, true by definition, statements are the primary intended meaning of the disposition analogy / metaphor. The relational model definitely does not limit inquiry in this particular way. <br />Of course, something about the way I worded all this is unfair to the relational approach. I phrased it as if the relational approach was something that came after the TSM approach, when I am quite confident that Gibson was doing the relational thing years earlier than the TSM papers.Eric Charleshttps://www.blogger.com/profile/17412168482569793996noreply@blogger.comtag:blogger.com,1999:blog-9192597712746432631.post-54146551798607004172011-04-10T09:42:04.861+01:002011-04-10T09:42:04.861+01:00You know, I don't think I've ever read the...You know, I don't think I've ever read the Cutting paper. I should look.<br /><br />Given you agree with some of the critique here, why do you think relations are still the way to go?Andrewhttps://www.blogger.com/profile/16732977871048876430noreply@blogger.comtag:blogger.com,1999:blog-9192597712746432631.post-29751428923605121202011-04-09T22:11:44.567+01:002011-04-09T22:11:44.567+01:00A few quick comments:
1) I adamantly believe the r...A few quick comments:<br />1) I adamantly believe the relational view of affordances is a better way to go, but I also agree with you that it is not NECESSARY to do many of the things Chemero wants. Nice analysis. <br />2) I like your coverage of the problem of two minds in the other post. Indeed, that was a central issue for James, Holt, and Perry (Perry was another chief in the New Realism movement, and was James’s first major biographer).<br />3) One odd thing about Chemero calling relations “Affordance 2.0” is that they were also “Affordances 0.5” That is, I think Gibson was clearly thought of affordances as relational. He approved of what TSM were doing, but I think Cutting was right that the two lines of thinking differ significantly (do you know the 2 ecological psychologies paper? More obvious now than at the time.).Eric Charleshttps://www.blogger.com/profile/17412168482569793996noreply@blogger.com