Re: A purely relational ontology?

[Brent Meeker]

 Good.  I like it.

On 6/17/2019 7:15 PM, Pierz wrote:

I've been thinking and writing a lot recently about  a conception of reality which avoids the debates about what is fundamental in reality. It seems to me that with regards to materialism, we find it very difficult to escape the evolutionarily evolved, inbuilt notion of "things" and "stuff" that our brains need in order to manipulate the world. Yet QM and importantly the expected dissolution of time and space as fundamental entities in physics have made any such simple mechanistic notion of matter obsolete - what is left of matter except mathematics and some strange thing we can only call "instantiation" - the fact that things have specific values rather than (seeming to be) pure abstractions? What does a sophisticated materialist today place his or her faith in exactly? Something along the lines of the idea that the world is fundamentally describable by mathematics, impersonal and reducible to the operation of its simplest components. With regards to the last part - reductionism - that also seems to be hitting a limit in the sense that, while we have some supposed candidates for fundamental entities (whether quantum fields, branes or whatever), there is always a problem with anything considered "fundamental" - namely the old turtle stack problem. If the world is really made of any fundamental entity, then fundamentally it is made of magic - since the properties of that fundamental thing must simply be given rather than depending on some other set of relations. While physicists on the one hand continually search for such an entity, on the other they immediately reject any candidate as soon as it is found, since the question naturally arises, why this way and not that? What do these properties depend on? Furthermore, the fine tuning problem, unless it can be solved by proof that the world *has* to be the way it is – a forlorn hope it seems to me – suggests that the idea that we can explain all of reality in terms of the analysis of parts (emergent relationships) is likely to collapse – we will need to invoke a cosmological context in order to explain the behaviour of the parts. It's no wonder so many physicists hate that idea, since it runs against the deep reductionist grain. And after all, analysis of emergent relationships (the parts of a thing) is always so much easier than analysis of contextual relationships (what a thing is part of). 

To get to the point then, I am considering the idea of a purely relational ontology, one in which all that exists are relationships. There are no entities with intrinsic properties, but only a web of relational properties. Entities with intrinsic properties are necessary components of any finite, bounded theory, and in fact such entities form the boundaries of the theory, the "approximations" it necessarily invokes in order to draw a line somewhere in the potentially unbounded phenomenological field. In economic theory for instance, we have “rational, self-interested” agents invoked as fundamental entities with rationality and self-interest deemed intrinsic, even though clearly such properties are, in reality, relational properties that depend on evolutionary and psychological factors, that, when analysed, reveal the inaccuracies and approximations of that theory. I am claiming that all properties imagined as intrinsic are approximations of this sort - ultimately to be revealed as derived from relations either external or internal to that entity.

Of course, a purely relational ontology necessarily involves an infinite regress of relationships, but it seems to me that we must choose our poison here - the magic of intrinsic properties, or the infinite regress of only relational ones. I prefer the latter. (Note that I am using a definition of relational properties that includes emergent properties as relational, though the traditional philosophical use of those terms probably would not. The reason is that I am interested in what is ontologically intrinsic, not semantically intrinsic.)

I wonder what you think of the observation that epistemology precedes ontology.  Epistemology is relational; it's about  how you know about that.  I tend to view it as more basic than ontology.  Our observations are much more stable than the ontology of our theories.    That's one of things makes me skeptical of Bruno's theories. The fact that he discounts observation as a source of knowledge (mere Aristotelianism) and wants to derive everything from abstract reasoning and equate knowledge with provability.

What would such a conception imply in the philosophy of mind? Traditionally, the “qualiophiles” have defined qualia as intrinsic properties, yet (while I am no fan of eliminativism) I think Dennett has made a strong case against this idea. Qualia appear to me to be properties of relationships between organisms and their environments.

And qualia are identified with consciousness.  Yet it seems that conscious thoughts and feelings are often best explained by somatic events that are subconscious.

Brent

They are not fundamental, but then neither is the “stuff” of which organisms and environments are made. We simply cannot ask about fundamental properties, but must confine ourselves to the networks of relationships we find ourselves embedded in, and from which we, as observer-participants, cannot be extricated. “Third person” accounts, including physics, are abstractions from aggregations of first person accounts, and none can rise so high above the field of observation as to entirely transcend their origins in the first person. Thus there are certainly objective truths, but not Objective Truths, that is truths that are entirely unbound to any observer and which nominate the absolute properties of real objective things.

Note that the “relationalism” I am proposing does not in any way imply *relativism*, which flattens out truth claims at the level of culture. Nor does it make consciousness “primary”, or mathematics. I cannot personally reconcile the interior views (qualia, if you like, though I think that terms places an unwarranted emphasis on “what experiences are like” rather than the mere fact of experience) with a purely mathematical ontology.

One obvious objection to this whole idea is the counter-intuitiveness of the idea of relationships without “things” being related. Yet I think the fault lies with intuition here. Western thinking is deeply intellectually addicted to the notion of “things”. David Mermin has interpreted QM in terms of “correlations only” – correlations without correlata as he puts it – an application of similar ideas to quantum theory. Part of the objection I think lies in the semantics of the word “relationship”, which automatically causes us to imagine two things on either side of the relation. It would be better to think in terms of a web, then, than individual, related entities. Or simply say that the related entities are themselves sets of relationships. Mathematics provides a good example of such a purely relational domain – a number exists solely by virtue of its relationships with other numbers. It has no intrinsic properties.

Yet what then of the problem of specific values – the instantiation aspect of materialism? To quote Hedda Mørch:  “… physical structure must be realized or implemented by some stuff or substance that is itself not purely structural. Otherwise, there would be no clear difference between physical and mere mathematical structure, or between the concrete universe and a mere abstraction.”

We can overcome such an objection by invoking the first person perspective. Mørch credits the specific values of entities in our environment (some specific electron having this position, that momentum and so on) to some property of “being instantiated in something intrinsic”, harking back to Kant’s Ding an Sich. Yet there is an alternative way of viewing the situation.  Let us imagine that each integer was conscious and able to survey its context in the field of all numbers. Take some number, let us say 7965. When number 7965 looks around, it sees the number 7964 right behind it, and the number 7966 right ahead. Trying to understand itself and the nature of its world, it starts doing arithmetic and finds that everything  around it can be understood purely in terms of relational properties. Yet it says to itself, how can this be? Why do the numbers around me have the specific values they do? What “breathes fire” into those arithmetical relations to instantiate the specific world I see? Yet 7965 is wrong. It is ignoring the significance of the first-person relation that places it within a specific context that defines both it and the world it sees.

Note that I am not, like Bruno, actually suggesting that numbers are conscious. The point of the thought experiment is merely to show how specific values can exist within a first person account, without us needing to invoke some unknowable thing-in-itself or substrate of intrinsic properties. 

Grateful for any comments/critiques.

--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/868bd041-299a-4618-9586-4b6362755cd7%40googlegroups.com.

[Bruno Marchal]

On 18 Jun 2019, at 20:12, 'Brent Meeker' via Everything List <everyth...@googlegroups.com> wrote:

 Good.  I like it.

On 6/17/2019 7:15 PM, Pierz wrote:

I've been thinking and writing a lot recently about  a conception of reality which avoids the debates about what is fundamental in reality. It seems to me that with regards to materialism, we find it very difficult to escape the evolutionarily evolved, inbuilt notion of "things" and "stuff" that our brains need in order to manipulate the world. Yet QM and importantly the expected dissolution of time and space as fundamental entities in physics have made any such simple mechanistic notion of matter obsolete - what is left of matter except mathematics and some strange thing we can only call "instantiation" - the fact that things have specific values rather than (seeming to be) pure abstractions? What does a sophisticated materialist today place his or her faith in exactly? Something along the lines of the idea that the world is fundamentally describable by mathematics, impersonal and reducible to the operation of its simplest components. With regards to the last part - reductionism - that also seems to be hitting a limit in the sense that, while we have some supposed candidates for fundamental entities (whether quantum fields, branes or whatever), there is always a problem with anything considered "fundamental" - namely the old turtle stack problem. If the world is really made of any fundamental entity, then fundamentally it is made of magic - since the properties of that fundamental thing must simply be given rather than depending on some other set of relations. While physicists on the one hand continually search for such an entity, on the other they immediately reject any candidate as soon as it is found, since the question naturally arises, why this way and not that? What do these properties depend on? Furthermore, the fine tuning problem, unless it can be solved by proof that the world *has* to be the way it is – a forlorn hope it seems to me – suggests that the idea that we can explain all of reality in terms of the analysis of parts (emergent relationships) is likely to collapse – we will need to invoke a cosmological context in order to explain the behaviour of the parts. It's no wonder so many physicists hate that idea, since it runs against the deep reductionist grain. And after all, analysis of emergent relationships (the parts of a thing) is always so much easier than analysis of contextual relationships (what a thing is part of). 

To get to the point then, I am considering the idea of a purely relational ontology, one in which all that exists are relationships. There are no entities with intrinsic properties, but only a web of relational properties. Entities with intrinsic properties are necessary components of any finite, bounded theory, and in fact such entities form the boundaries of the theory, the "approximations" it necessarily invokes in order to draw a line somewhere in the potentially unbounded phenomenological field. In economic theory for instance, we have “rational, self-interested” agents invoked as fundamental entities with rationality and self-interest deemed intrinsic, even though clearly such properties are, in reality, relational properties that depend on evolutionary and psychological factors, that, when analysed, reveal the inaccuracies and approximations of that theory. I am claiming that all properties imagined as intrinsic are approximations of this sort - ultimately to be revealed as derived from relations either external or internal to that entity.

Of course, a purely relational ontology necessarily involves an infinite regress of relationships, but it seems to me that we must choose our poison here - the magic of intrinsic properties, or the infinite regress of only relational ones. I prefer the latter. (Note that I am using a definition of relational properties that includes emergent properties as relational, though the traditional philosophical use of those terms probably would not. The reason is that I am interested in what is ontologically intrinsic, not semantically intrinsic.)

I wonder what you think of the observation that epistemology precedes ontology.  Epistemology is relational; it's about  how you know about that.  I tend to view it as more basic than ontology.  Our observations are much more stable than the ontology of our theories.    That's one of things makes me skeptical of Bruno's theories. The fact that he discounts observation as a source of knowledge (mere Aristotelianism) and wants to derive everything from abstract reasoning and equate knowledge with provability.

Mechanism typically distinguish provability and knowledge.

For provability we don’t have []p -> p as a theorem, but we have it for knowledge, and this is what makes sense full to *define*, following Theaetetus, knowledge by true provability. It obeys S4 (the traditional modal logic of knowledge).

Silarly, it is plain false that I dismiss observation as a source of important information, and even as the only mean to refute any theory bearing on some “reality”. I would not have concsecrated 35 years to show that Mechanism is testable empirically, and test it, if that was not the case. Again, incompleteness distinguishes observability from knowledge and provability. It obeys a version of B, the modal logic corresponding to quantum logic.

What would such a conception imply in the philosophy of mind? Traditionally, the “qualiophiles” have defined qualia as intrinsic properties, yet (while I am no fan of eliminativism) I think Dennett has made a strong case against this idea. Qualia appear to me to be properties of relationships between organisms and their environments.

And qualia are identified with consciousness.  Yet it seems that conscious thoughts and feelings are often best explained by somatic events that are subconscious.

Any qualia necessitates consciousness, but consciousness does not necessitate much of qualia. With some drugs, we can loose all qualia but one, which is sometimes described as pure consciousness. Consciousness is just a more general term for qualia. 

Yes, all this related to somatic events, which eventually are related to infinitely many number relations.

Bruno 

Brent

They are not fundamental, but then neither is the “stuff” of which organisms and environments are made. We simply cannot ask about fundamental properties, but must confine ourselves to the networks of relationships we find ourselves embedded in, and from which we, as observer-participants, cannot be extricated. “Third person” accounts, including physics, are abstractions from aggregations of first person accounts, and none can rise so high above the field of observation as to entirely transcend their origins in the first person. Thus there are certainly objective truths, but not Objective Truths, that is truths that are entirely unbound to any observer and which nominate the absolute properties of real objective things.

Note that the “relationalism” I am proposing does not in any way imply *relativism*, which flattens out truth claims at the level of culture. Nor does it make consciousness “primary”, or mathematics. I cannot personally reconcile the interior views (qualia, if you like, though I think that terms places an unwarranted emphasis on “what experiences are like” rather than the mere fact of experience) with a purely mathematical ontology.

One obvious objection to this whole idea is the counter-intuitiveness of the idea of relationships without “things” being related. Yet I think the fault lies with intuition here. Western thinking is deeply intellectually addicted to the notion of “things”. David Mermin has interpreted QM in terms of “correlations only” – correlations without correlata as he puts it – an application of similar ideas to quantum theory. Part of the objection I think lies in the semantics of the word “relationship”, which automatically causes us to imagine two things on either side of the relation. It would be better to think in terms of a web, then, than individual, related entities. Or simply say that the related entities are themselves sets of relationships. Mathematics provides a good example of such a purely relational domain – a number exists solely by virtue of its relationships with other numbers. It has no intrinsic properties.

Yet what then of the problem of specific values – the instantiation aspect of materialism? To quote Hedda Mørch:  “… physical structure must be realized or implemented by some stuff or substance that is itself not purely structural. Otherwise, there would be no clear difference between physical and mere mathematical structure, or between the concrete universe and a mere abstraction.”

We can overcome such an objection by invoking the first person perspective. Mørch credits the specific values of entities in our environment (some specific electron having this position, that momentum and so on) to some property of “being instantiated in something intrinsic”, harking back to Kant’s Ding an Sich. Yet there is an alternative way of viewing the situation.  Let us imagine that each integer was conscious and able to survey its context in the field of all numbers. Take some number, let us say 7965. When number 7965 looks around, it sees the number 7964 right behind it, and the number 7966 right ahead. Trying to understand itself and the nature of its world, it starts doing arithmetic and finds that everything  around it can be understood purely in terms of relational properties. Yet it says to itself, how can this be? Why do the numbers around me have the specific values they do? What “breathes fire” into those arithmetical relations to instantiate the specific world I see? Yet 7965 is wrong. It is ignoring the significance of the first-person relation that places it within a specific context that defines both it and the world it sees.

Note that I am not, like Bruno, actually suggesting that numbers are conscious. The point of the thought experiment is merely to show how specific values can exist within a first person account, without us needing to invoke some unknowable thing-in-itself or substrate of intrinsic properties. 

Grateful for any comments/critiques.

--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/868bd041-299a-4618-9586-4b6362755cd7%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/981c39ee-353e-1ea3-43ad-322be386f7f9%40verizon.net.