Psyche: Attac on Receptivity (Robinson and McKitrick)

[Tychic]

There is some discussion of Rosenbergs „A Place for Consciousness“ (APFC) in the December issue of Psyche. All articles are available in the World Wide Web. See
http://psyche.cs.monash.edu.au/"

1) William S. Robinson has some „Doubts on Receptivity“. It may be only a trivially truth that effectiveness and receptivity are logical compliments of one another. There is no new idea in this as there is in the logical relation of Northness and Southness.

To understand what I mean by the triviality problem, imagine that someone announces the discovery of a logical relation between Northness and Southness, and supports the claim to this discovery by saying that the world cannot contain a case of something’s being North of something else unless it also contains a case of something’s being South of something else. I hope and believe that most readers will regard such a claim as an invitation to deflation. “B is South of A” does not introduce a new fact, it is just an alternative way of saying that A is North of B.

An obvious difference is that for Rosenberg receptive properties are connective relations while effective properties can be one-place. As far as I see the duality between law of nature and disposition is far more similar to the conceptual connection between receptivity and effectiveness. Rosenberg himself calls the effective properties "effective dispositions". He refrains to call his receptive properties laws of nature or instances of laws of nature, but at least in his example of a toy Newtonian physics the level two individuals correspond to instances of Newton’s laws. I'm sure that both not each regularity of nature corresponds to law of nature and not all laws of nature correspond abstractions of receptive properties. But I suppose that each receptive property corresponds to at least one law of nature. (See Rosenberg's hierarchy of Humean, nomic and causal mesh of the facts. Each mesh includes and refines the previous. See chapter 11 of APC.)
This does not solve Robinson's doubts since one of the two concepts. Many think that dispositions are reducible to laws of nature, or vice versa. If it is like that, they cannot be two sides of the wall of causality.
In the second part of his essay Robinson recognizes the binding function of receptive properties: Receptive properties constitute individuals. But this only helps to reformulate his triviality problem in another way. You may collect some effective connections of some individuals and rename them as common receptivity. The question remains.

[…] we need an argument that the constituted individuals have any properties that are not results of the efficacy of the effective properties; and this argument does not seem to me to have been given.”

There is no positive argument as required by Robinson in APFC. Rosenberg proposes a model how this could be possible. His idea is that the constraints of a level N receptivity cannot be reconstructed by level N individuals and their effective properties. Given only these effective properties there is too much undetermined. No receptive connection without real constraint to the lower level effective properties. (One may call this an emergence of a new individual (of level N+1 or not. The counterexample mentioned by Robinson would not work with this kind of emergence.)

2) Jennifer McKitrick gives a more detailed analysis and criticism in „Rosenberg on Causation“. Dispositions are for McKitrick intrinsic properties of individuals and may exist even if a disposition is never manifested. For a relation to exist there have to be at least two objects. Therefore, a relation cannot be a dispositional property. But Rosenberg calls his receptive properties on the one side connective and relational properties and on the other side receptive dispositions.
- I think Rosenberg could accept McKitrick’s supposition concerning the intrinsicallity of dispositions: Again the hierarchical structure of individuals may provide some help. Insofar the receptivity of the many level N individuals is on focus no one can see how their shared receptive property may be dispositional. Of course it is not intrinsic to anyone of these level N individuals but intrinsic to them. But it may be an intrinsic and dispositional property of the new constituted level N+1 individual.

3) McKitrick dislikes the idea of receptive powers as connections. She does not see (or at least she does not value) its positive contribution on the Theory of Causal Significance. Receptive properties seem to be beyond the real of scientific investigation. She accepts that they have to be introduced on conceptual reasons. But why not prefer the simple solution without relational properties?
– I try to answer: Rosenberg may say that this would be a trivial understanding of receptivity (to use the terms of Robinson). This trivial understanding does not correspond to logical necessity of a concept of receptivity. Besides, Rosenberg perhaps could resist on having the really simpler concept of causality.

4) McKitrick further thinks that receptive properties that are needed to bind effective properties are faced to a version of the “third man problem”.
- I answer: Again the hierarchy of individuals may stop the regress. If there is nothing further to constrain and the reality is determinate there is no further receptive property necessary to bind anything.

5) The hierarchical structure of individuals is a big problem for McKitrick. On Rosenberg’s account individuals of lower levels that are parts of an individual of a higher level are not actual as long as the higher level individual constraints them. This makes problems (i) conceiving of binding and causal activity in non actual realms and (ii) for the individuation of properties.
Ad (i): Conceiving non actual entities as conceiving actual entities independent of their environment does not very much, as McKitrick suggests. This puts them only from real space to possibility space, from one environment into another.
– But why not, I would ask, take non actual individuals as unconstrained individuals as Rosenberg suggests. A kidney in the hand of a surgeon one second before he transplants it to a new body is not constraint to clean blood because it not constrained by the rest of an human body to do its proper function. (This model says: There is no actual kidney in the time before the transplantation but only a possible kidney.) The talk non actuality as considered independent of environment can be refined by saying: independent of the constraints of higher level individuals.
Ad (ii): Suppose there is a positive charge in the actual world. The following counterfactual would be true in Rosenberg’s theory:

If this instance of positive charge were not bound to other properties, it would have been an indeterminate charge.

This makes problems with the identity conditions for properties

It’s not obvious that an instance of having some charge or other can be the same property as an instance of positive charge. That possibility violates Leibniz’s Law of Indiscernability of Identicals: If property instance P = property instance Q, then if P is F, then Q is F. It seems essential to the instance of indeterminate charge that it is indeterminate and essential to the instance of positive charge that it is positive.

- I think like McKitrick that a haecceitas of properties will not be a good candidate for the identity conditions. I do not think that there is a relation of identity between the determinate and the determinable property. I do not really see McKitrick’s problem. I would say

If this instance of positive charge were not bound to other properties, it would not be an instance at all.​

6) In section 4.1 McKitrick supposes that effective dispositions are of no need of a carrier. Her own theory of causation and the attempts of Mumford and Mulnar also make this point. Unfortunately, McKitrick does not introduce these theories and I do not know them. In addition, she does not show how these theories disprove Rosenberg’s argument for the carrier theory.

 

[TakLoufer]

There is some discussion of Rosenbergs „A Place for Consciousness“ (APFC) in the December issue of Psyche. All articles are available in the World Wide Web. See
http://psyche.cs.monash.edu.au/

1) William S. Robinson has some „Doubts on Receptivity“. It may be only a trivially truth that effectiveness and receptivity are logical compliments of one another. There is no new idea in this as there is in the logical relation of Northness and Southness.


An obvious difference is that for Rosenberg receptive properties are connective relations while effective properties can be one-place. As far as I see the duality between law of nature and disposition is far more similar to the conceptual connection between receptivity and effectiveness. Rosenberg himself calls the effective properties "effective dispositions". He refrains to call his receptive properties laws of nature or instances of laws of nature, but at least in his example of a toy Newtonian physics the level two individuals correspond to instances of Newton’s laws. I'm sure that both not each regularity of nature corresponds to law of nature and not all laws of nature correspond abstractions of receptive properties. But I suppose that each receptive property corresponds to at least one law of nature. (See Rosenberg's hierarchy of Humean, nomic and causal mesh of the facts. Each mesh includes and refines the previous. See chapter 11 of APC.)
This does not solve Robinson's doubts since one of the two concepts. Many think that dispositions are reducible to laws of nature, or vice versa. If it is like that, they cannot be two sides of the wall of causality.
In the second part of his essay Robinson recognizes the binding function of receptive properties: Receptive properties constitute individuals. But this only helps to reformulate his triviality problem in another way. You may collect some effective connections of some individuals and rename them as common receptivity. The question remains.



There is no positive argument as required by Robinson in APFC. Rosenberg proposes a model how this could be possible. His idea is that the constraints of a level N receptivity cannot be reconstructed by level N individuals and their effective properties. Given only these effective properties there is too much undetermined. No receptive connection without real constraint to the lower level effective properties. (One may call this an emergence of a new individual (of level N+1 or not. The counterexample mentioned by Robinson would not work with this kind of emergence.)

Strange how no one followed up on this post. Where's Hypnagogue? He seems to have a knack for explaining Rosenberg's ideas better than Rosenberg himself. Oh well, I suppose I'll give it a shot. Regarding Robinson's criticism, I'll add that I too have always found Rosenberg's receptivity somewhat . . . well, unnecessary, perhaps. It all fits wonderfully and gives a good explanation for consciousness/subjectivity, but, consciousness aside, the whole concept doesn't seem particularly needed. One could just as easily have all the work done by the effective side.

Effective properties would, by their very definition, have effects on the rest of the world (i.e. other effective properties). Natural individuals? Well, different effective properties would have certain dispositions to "link up" under certain circumstances into individuals with a shared responsiveness. Their nature would also include the emergent properties associated with higher level individuals. This alone should account for Rosenberg's natural individuals without needing a second and distinct "receptive" property.

Or would it?

I admit I am confused about all this, and I believe my confusion may stem from my habit of thinking of the effective properties as objects -- like little "charges" and "spins" and "colors" floating about in a pre-existing void, like the components of a cosmic alphabet soup. Envisioning them this way, it seems logical to assume they could effect each other without a receptive property as a mediator, much the same way ordinary objects effect and influence each other. In short, I'm visualizing the model in the form of everyday physics. To do otherwise seems counter-intuitive.

But then, I know that this is a naive way of thinking. Effective properties are not objects; they are properties that are, well, effective. That's all they are. They can effect other things (i.e. receptive properties), and nothing else. By thinking that they can themselves be affected is to grant them more reality then they have.

If a collection of effective properties were to "link up" into a natural individual, one has to ask, 'what is being linked up?' Effective properties alone would seem to lack any concrete nature that would allow another EP to absorb it. After all, what's being absorbed into the natural individual? If it's the EP itself, then the EP evidently possesses a nature above and beyond the basic power to effect . . . it must have a property that is being effected.

Without a receptive component, one would be in the position of asking: what is there about Effective Property B that Effective Property A can influence? What is EP A effecting about EP B? If it's EP B's "effectiveness," then this effective nature has another facet -- the power to be effected . . . or in other words, it has receptivity.

For example, let's say there are two EPs. One is carried by the phenomenal color red, the other by blue. Unlike what Robinson proposes, these phenomenal qualities are not experiential. No subject experieces them, so in a way the colors don't really exist, or at least they only potentially exist. With this being said, in the absence of receptivity (i.e. an experiencer) can the EP red effect the EP blue? It doesn't seem that they could, since a color unexperienced doesn't really exist. They would be nothing to each other.

Now, Robinson proposes that phenomenal qualities may be the same as experiential qualities, but this doesn't make much sense to me; they are both distinct conceptually, though they are interdependent. But, if what he says is true, then the EPs of red and blue would also have the additional character of being experienced, in whcih case it would make sense for EP red to effect EP blue and vice versa, since the the two EPs would have a more concrete reality. The natural individual of EP-red + EP-blue would be carried by a subject that would presumably experience the color purple, and this subject would emerge from the union of the two EPs -- each of which possesses it's own receptivity/experience. Even still, Robinson's alternative doesn't dispose of receptivity, it just implicates it into the EP itself, rendering it monadic.

If we follow this reasoning, by necessity all EFs would also be receptive. I suppose we can call them effective-receptive properties (which would make them for all intents and purposes objects) . . . but this goes back into the monadic problem. If we're going to posit receptivity into EPs, we would do well to have it be a connective property to allow the global nature of natural individuals, which is exactly what Rosenberg proposes.

I think this reasoning somewhat answers Robinson's criticisms, but I hope someone responds to clear things up a bit.

 

[Tychic]

Receptivity: gainst epiphenomenalism and manifestation of levels of nature

Hi TakLoufer, thank you for joining the forum and for your interesting response to my post. (I also like to think of properties like objects )

Effective properties would, by their very definition, have effects on the rest of the world (i.e. other effective properties). Natural individuals? Well, different effective properties would have certain dispositions to "link up" under certain circumstances into individuals with a shared responsiveness. Their nature would also include the emergent properties associated with higher level individuals. This alone should account for Rosenberg's natural individuals without needing a second and distinct "receptive" property.

I agree and disagree that the need of receptivity is no more than experience. Receptivity is introduced against the charge of epiphenomenalism. It opens the possibility to introduce downward causation for experiential properties – without contradiction to the causal closure of the physical. I agree with you that the only macroscopic entity that is of urgent need of downward causation is consciousness/experience.

My disagreement: It is a natural idea to suppose that there are other things that are causal agents on themselves: atoms, molecules, cells etc. If there is a possibility for this without denying the closure of the physics this is truly welcome. (Science has no answer why some molecules have the ‘idea’ to built up and act as a protozoa.)

If a collection of effective properties were to "link up" into a natural individual, one has to ask, 'what is being linked up?' Effective properties alone would seem to lack any concrete nature that would allow another EP to absorb it. After all, what's being absorbed into the natural individual? If it's the EP itself, then the EP evidently possesses a nature above and beyond the basic power to effect . . . it must have a property that is being effected.

I disagree: some effective properties are concrete, some others are abstract. They are concrete insofar as they are determinate, and they are abstract insofar as they are determinable. (Concrete effective properties can only be an effective cause in temporal successive causal nexus. If you suppose with Rosenberg that simultaneous causation is possible the “causes” cannot be fully concrete.)
A second disagreement: As far as I see effective properties can be absorbed from other effective properties without a receptive property coming into play. Only if there is some absorting that is not constraint by the causal laws of the same level of nature receptive connections come into play.
As far as I see receptivity is introduced to make sense of levels of nature.