Greg Detre
Sunday, October 20, 2002
"�[to
find out what else] is needed for a complete
picture of intentionality, semantics and ontology ...[for which one needs] to
understand how a conception of objects can arise on a substrate of infinitely
extensive fields of particularity" [p. 191]. � from the Vellino review
�perception is an activity on the part of an intentional subject, and furthermore, an activity that relates it to, or engages it with, something in the [typically external] world�
�we talk of perceiving a sparrow, a wry look, a distant thundercloud � but not, except on higher-order reflection, of perceiving our perceptions or sensations�
it also �implies a kind of intentional success, an achievement�. To say that a sailor perceives an island on the horizon implies, unless otherwise noted, that there was an island to be perceived, and that the sialor did indeed see it. Moreover, as I will put it, it is the island itself, not an impression of the island, that figures in the content of the sailor�s intentional stare�
he�s not denying that there�s a �complex causal relation between the sailor and island, perhaps mediated by light waves and binoculars � But none of these things is what perception actually is � they relate the subject to the wrong thing�
the grammar of �perception� reflects that:
reaches out into the world � direct object
it is (by default) veridical � �
�register� is much broader than �perception�
��she registered a table� will be taken to imply not only that there was a table to be registered, but that she successfully ended up in an intentional relationship to it�
he wants registration to �include all intentional actions that can legitimately be said to have some form of content�
isn�t that tautological???
�it will at least include cognition, sensing, having intuitions about, and a number of other things � and will perhaps include willed action, action that assumes that the world is given or structured in some way�
how is registration different from simply �intentional thought�???
��to register the world, that is, is to do or be oriented towards the world in such a way that it presents or arranges or constitutes itself as world�
pah, that�s pretty meaningless
�in
the special case of objects, �register� can be taken as being related in
meanign to �individuate� or �identify�, but a much closer synonym is
�objectify�, which �� ???
�I will take registration to include quality, mass substances � even feature-placing, so long as there is reason to believe that the features are placed in a world�
�register� � connotations from ordinary English:
a salutary sharing of responsibility for a subject�s taking an object to be an object � but responsibility for intentional success falls to some degree on both registrar and registered
registering as something � sense of �aspectual shape� or �intensionality� characteristic of all intentional activity
alignment with the external situation � residual sense of realism
sense of involvement/engagement with the world, even calibration
reminder that �the ways in which we take the world to be are in part a function of the assumptions and culture and machinery, including conceptual machinery, with which we approach it�
rather than �conceptual scheme�, he prefers to say that we talk/deal with the world in different registers
register � 3 essential properties:
registration is the net activity that leads to a conception of, or take on, or intentional attitude towards, the world as given/available � as the world
originally neutral as to the subject/world split
does not single out objects as a premier ontological category/class
3 properties initially identified as characteristic of objects:
objectivity
particularity
individuality
physics can supply a world-extensive cosmic particularity, but can�t help with individuality
how could a notion of individuality arise on a substrate of fields of particularity?
foreswore a sharp subject/object distinction along the way � in line with the intuition that objects emerge collaboratively from subject-object interaction
so the question becomes:
how does registration arise on a substrate of infinitely extensive fields of pure particularity
how, and in what ways, is the
registration of objects a distinct species?
assembling a positive picture
observation:
�the multi-various parts of the world do not march in lock-step together�
�the world is fundamentally characterised by an underlying flex or slop � a kind of slack or �play� that allows some bits to move about or adjust without much influencing, and without being much influenced by, other bits�
contrasts this with a world consisting of nothing but an endless series of interlocked gears
�the gear world would lack slop � effects would not dissipate � if one gear were to move by even a tiny amount, every other gear in the universe, no matter how far flung, would instantly and proportionally be affected�
even if you consider the �butterfly effect� in our world, it�s only remotely possible, and such effects take time
it�s still true that the �far and away the majority of micro-disturbances quickly die away. In our world, especially to the extent that we find it coherent, effects by and large dissipate�
the world�s primordial flex/play does two crucial things;
establishes the problem that intentionality solves
provides the wherewithal for its solution
effective reach:
�because of the dissipative nature of the playing field, an enduring entity cannot, at any given moment, be affected by things that, at that same moment, are beyond what I will call effective reach�
it�s �essentially the gradual falling off or dissipation of influence familiar from physics�
�there comes a point where the effective illuminative coupling [the lights from San Francisco seen from a distance through fog] subsides below the level at which it can any longer do any work � [though] San Francisco still matters because of past connections, present interests, future possibilities�
intentional directedness
�intentional directedness is not held in place with physical glue. Long after the phsyical tie has been broken, or has stretched too thin to hold anythhing together � the semantic relation tying a subject to the subject matter will persevere ��
simple but effective strategy:
�stay oriented towards what is no longer effectively discriminable�
describes the super-sunflower
what�s special is that �they track something to which they are not effectively coupled� (non-effective tracking)
�the forerunner of semantics
a �very simple form of effect-transcending coordination in some way essential to the overall existence or well-being of the constituted system�
�Separation means flex, in other words; the point of the arrangement is to allow the data, at least potentially, to flop around under some external influence without disturbing the code. Then, when the word is read, what was temporarily separated is brought back into contact in order to ensure that the coordination between the two has been maintained�
�There is nothing more basic to intentionality than this pattern of coming together and coming apart, at one moment being fully engaged, at another point being separated, but separated � this is the point � in such a way as to stay cooirdinate with what at the moment is distal and beyond effective reach�
flex and slop underwrite the very notion of connection, disconnection and the limits of effective reach
otherwise, there�d probably be no warrant for saying the world had parts at all
because of the slop, it�s not automatic that the proximal system will stay synchronised with what is distal
this is why there is a �sense in which slop �establishes the problem of intentionality��
�it�s a very important consequence of this slop that the proximal system can relatively easily adjust its own state without having to drag much of the surrounding territory with it�
the coordinating regularity will become the semantic
relation
the coordinated-with regularity will become the type structure of the referent
there has to be a balance (�intermediately flexible�) between too much flex and and too little
�the word �play� is sometimes useful to describe both the world�s flex and the intentional behaviour that deals with it � i.e. � something crucially intermediate between chaos and rigidity�
�it is as fundamental a fact as any about this metaphysics that it is based on an ineliminable notion of �playfulness� � a kind of irreducible, obstreperous, wily refusal ever to be formally captured and written down�
�Intentionality is a way of
exploiting local freedom or slop in order to establish coordination with what
is beyond effective reach�
registration, and intentionality in general, are not effective phenomena
intentionality, and objects in general (as a result???) are not causal/do not have a causal explanation
�If what it is to be an object is
in part to be constituted by regularities in which intentional subjects are
participants, and if intentionality requires distance and non-effective
coordination, then it follows that all stories about objects, and indeed all
stories in toto (since the point holds of registration, ultimately, not solely
of objects) will inherently trade, more or less explicitly, in non-effective
regularities�
�intentionality is not an effective phenomenon, and will not have an effective explanation, and similarly that objects are non-effective, and will not have effective explanations�
insofar as �causal� means �effective�, intentionality and objects are non-causal
�That semantic reach exceeds effective grasp is essentially a theorem of this metaphysical account�
�foresworn a sharp representation/ontology boundary�???
�� I am going to take registration to be the ultimate subject matter, and therefore will not swear allegiance to a sharp representational/ontological split, it follows that anything true of semantics � i.e. of intentionality � will also and indissolubly be true of objects�
�being an object cannot be a physically effective property, since there is no way in which a temporally extended object could have a physical effect distinct from that of a non-temporally-extended instantaneous time-slice of that object�
�this is yet another reason to support ch 5�s contention that physics is not ontologically committed to individual objects�
�abstraction, essential to the notion of an object, is like semantic reach in being physically transcendent�
tracking is not registration because it is not appropriately disconnected
however, �a property we ultimately associate with both subject and object � that of a boundary or edge � is in the first instance a property of their interaction�
�we tend to view the mediating air as conceptually as well as optically transparent�
�Gibson was too mild: it is not just that we do not represent the tree; we are not even distinct from the tree, at least not at first�
edge detectors
�there is a single common edge to a columnar-shaped multimedia disturbance that reaches continuously between (what we register as) frog and (what we register as) fly�
he prefers edge �participators� to �detectors� � like your sports jacket, which is connected to your motion (�it changes state in a way that is lawfully correlated with your motion�), but doesn�t really �detect� your motion
starts to use the s- and o-region terminology
when the coyote disappears from view (the causal loop/flow of effect is broken), �it is suddenly incumbent on the rabbit to take over responsibility for keeping �focused� on the coyote�
�an internal mechanism has to
compensate for what can no longer be relied on to be effectively provided by
the environment�
�� � this shouldering of effecitve responsibility by the s-region, to compensate for the break in effective coupling � is no less than the origin of reasoning, representation and syntax; the effective projection, onto the intentional agent, of the requisite arrangements for maintaining long-distance (semantic) coherence�
�as usual, what is true spatially is also true temporally. Because of physics� ahistoricity, any form of long-distance temporal coordination, either forward in time (prediction) or backwards in time (memory) will similarly involve a degree of disconnection�
�one must predict, which involves the same sort of retraction described in the text: a shouldering of effective resopnsibility by the s-region in order to develop or construct local and current (i.e. present) structures in standing in coordinated relation to phenomena that are still temporally remote�
�to register, the s-region must do more: it must play with the relation, even at the cost of a certain amount of destabilisation, so as to stabilise the far end. That is, the s-region must �sediment� or �extrude� the o-region as a discrete autonomous individual (at least partially distinct from the s-region) and thereby locate it in the wider world. It is the o-region itself, in other words, not the relation to the object, that must be �held� in at least partial stasis�
this requires detachment
�taking an object as an individual means gatering up an extended region of the flux and treating it as a unity�
�Note the intimate relation between the processes of abstraction and the aforementioned possibility of error implicit in the underlying slop�
�there are regularities � stabilities � that can only be established at these higher levels of abstraction�
abstraction �requires separation, in
order that the s-region not be buffeted by irrelevant details�
�registering [something] as a unity does not require that the thing itself be spatio-temporally or in any other way extensionless�
Haugeland: digitality is �a practical means to cope with the vagaries and vicissitudes, the noise and drift, of earthly existence�
deconvolve the deixis: I think this means (e.g. the acrobat with the flashlight) doing the opposite with one part of your body of what the rest of your body is doing, so that they cancel each other out
�the first way in which the self is gathered together as a unity may be as a stable locus of inversion�
�the first emergence of the �subject� as an individual would be as a long-term integral or aggregate of that which it must compensate for, in order to stabilise the rest of the world�
this is the �precursor to the later process of shifting the registration of the object from egocentric to allocentric coordinates. This helps it to begin the long and tough process of triangulating on the object, and washing out the contribution of everything else�.
http://www.ageofsig.org/people/bcsmith/index.html
Brian Cantwell Smith, �On the Origin of Objects�, 1996, MIT Press
From p. 333: "In fact one could think of the entire metaphysical project being argued in this book in the following way: as an answer to the question of what it would be like for everything to lie in a critical region, poised between regularity and chaos"
http://ontic.co.nz/Science/ObjectMind.htm
How
can we explain these concepts, computation, intentionality and representation
in terms of the physical? Brian Cantwell Smith in his 1996 publication On the
Origin of Objects proposes the following. In formulating a metaphysically safe
theory of computation we somehow need to "naturalise" computation.
And if we are to remain loyal to any of the raft of ideas forwarded by computational
theories of mind then naturalising computation means naturalising
intentionality and representation. Naturalistic philosophers take the
underlying natural world to be of the physical realm. This allows us to talk
about and develop metaphysically unproblematic theories about ordinary material
objects such as tables, chairs, rocks etc. Talking about activities of the
intentional realm, such as intentionality, mental representation, language,
content, psychology and the like are contrastingly metaphysically problematic
and naturalisation is the attempt to reduce or explain the latter in terms of
the former.
part of what I liked best about smith�s approach is its unabashed holism, to use a word that�s usually best avoided
presumably (in response to Vellino�s review of the book), the explanatory theory for individuals (rather than particulars) is (cognitive, or some other form of) psychology
we have to be prepared to forgive Smith the occasional lapse into what sounds somewhat like Eastern mysticism, as when he says, e.g.: �it is exactly by letting go while retaining appropriate directedness and orientation that the world �comes into presence��
to discuss Smith, we�re going to have unpack a lot of notions that get employed a lot, especially in philosophy, despite the fact that, and partly because, they�re really difficult to unpack
intentionality
reality
representation
ontology
subject/object
causation
isomorphism
particular vs individual (newish)
registration (new)
ontology - Ontology is the study of what there is, an inventory of what exists. An ontological commitment is a commitment to an existence claim.
intentionality - The property of the mind by which it is directed at, about, or 'of' objects and events in the world. Aboutness - in the manner of beliefs, fears, desires, etc.
intentio'nality n. the quality or fact of being intentional; esp. (in phenomenology) the fact of being directed at an object (as a supposed quality of every act of consciousness): E17.
intentionality, derived - The power of a system (e.g. the mind) to be "about" something if that power is derived from that system's connection to another, already intentional system. Language's intentionality is said to be derived from that of the mind.
semantics - The study of relations between a representation and what it represents
Theory of meaning; study of the signification of signs or symbols, as opposed to their formal relations (syntactics).
semantics, functional role - The meaning of a representation is the role of that representation in the cognitive life of the agent. It is an extension of the well known "use" theory of meaning as it supplements external use by including the role of a symbol inside a computer or a brain
semantics /sI"mantIks/ n.pl. Treated as sing. or (now rare) pl.L19. [f. prec.: see -ICS.]1 The branch of linguistics that deals with meaning; (the study or analysis of) the relationships between linguistic symbols and their meanings. L19. 2 transf. The interpretation of signs in general. M20.
intensionality - a context or form of words is intensional if its truth is dependent on the meanings, and not just the reference, of its component words
or on the meanings, and not just the truth-value, of any of its subclauses. So, 'He coughed because he smoked' is intensional, since there is no guarantee that truth is preserved if 'he smoked' is replaced by some other true sentence. More problematic are such contexts as 'The sales assistant thought that the customer was wrong', which supposedly may not be true if 'the customer' is replaced by the person's name, or by some other mode of reference, as 'The sales assistant thought that your cousin was wrong'. On the one hand, it has been maintained that we may enlist only referential terms which could have been used by the person whose thought is being reported, so that if the sales assistant was unaware that the customer was cousin to the person addressed, this second variant would be false. On the other hand, our ordinary practice would suggest that choice of referential terms is dictated more by what secures reference for those currently addressed, a correct mode of reference giving rise at worst to an inapposite form of words, not to a falsehood.
�Examples of particulars include the constitutive elemnets of what we ordinarily call (physical) objects (the physical cause of this patch of green/blue experience) but also events (such as Diana�s accident) and processes (the writing of this review). Contrast this with universals (e.g. the generic concept �canoe�) and abstract categories. The distinction between particulars and individals � and this is an important one � is that individuality is what makes an object discrete and separable from its background. An individual is consituted of particulars, but what makes it an individual is �� what allows us to say of one object that it is one; or two that they are two� (Smith)�.
�physics has something to say about particulars, but is silent about individuals�
criterion of ultimate correctness:
�No naturalistically palatable theory of intentionality-of mind, computation, semantics, ontology, objectivity-can presume the identity or existence of any individual object whatsoever.� (Smith)
�the world is neither entirely deterministic � a world of intertwined cogs and gears � nor is everything physically and causally disconnected from everything else�
�For it must be possible for me to register objects not only as present, here and now, in my perceptual field, but also as enduring, �stable� objects that preserve their identity somehow. This process of stabilisation requires the ability to deal with indexicals (such as this and now)��
�the process of individuation that takes place in selecting a �thing� from its environment and registering it as a thing (or object) requires an (intentional) act of separation or discrimination�
Smith:
�the world is depicted as one of cosmic and ultimately ineffable particularity�
�registration: a form of interaction, subsuming both representation and ontology, in which �s-regions� or subjects stabilise patches of the flux, in part through processes of intervention, adjusting and building them and beating them into shape, and also through patterns of disconnection and long-distance coordination, necessary in order to take the patch to be an object, or more generally to be something in and of the world� (pg 347
in contrast to how Vellino describes what Smith is saying, it does feel to me that to some degree Smith is reducing intentionality to something causal, though strictly speaking he�d say �effective��
but at the same time, that doesn�t feel quite right � because there�s something about the non-effective too � hmm, registration is kind of to do with the tracking when the subject and object (although I thought he wanted to get away from that dichotomy too???) are separated/independent enough
don�t forget that he starts talking about s- and o-regions � but how do we know their bounds??? presumably, by hindsight�
so is it kind of something about the way that the subject is arranged that it�s able to take responsibility for maintaining the connectedness with its object even while that connectiveness has been severed effectively???
I think that�s closer, and gets at the information processing aspect of intentionality, rather than some simple effective-causal-SR or isomorphism-representation bond�
is Smith�s philosophy of presence a plenum???
representation vs registration??? how big a notion is registration???
slop vs flex???
what�s the philosophy of computation in the book??? what is that??? how�s it relevant to the epistemology/metaphysics??? computation as reality??? cf Wolfram???
Husserl???
maybe re-read that context-aware article
so, is predictability/an internal model an essential part of registration???
the two are different � predictability can be a mechanical, intentional but non-representational process, whereas an internal model seems to be a much stronger notion
what about if I use Rolls� notion of a �representation� as basically being anything you can play around with � no, putting the distinction in the above way still works, because his definition defines a model, whereas mere predictability might govern the super-sunflower�s behaviour without necessarily being (internally, i.e. by the sunflower) manipulable
deictic vs indexical???
I think indexical = e.g. this/that, whereas deictic can include absolute (space/time) references/contexts�
in which sense of deictic is he using the term???
t-sentence???
particularity vs individuation???
�infinitely extensive fields of pure particularity�???
feature-placing???
what happens if you want to add a module that interacts with other modules in some unexpected way (for which there isn't a predefined message interface), or if you want to supercede some of the lower levels (e.g. to wander, and avoid everything but blue objects)???
subsumption vs inhibition???
post-Dartmouth
traditions in AI???
does he
really think that you can build all the way up to human-level intelligence in
this way???
To put it
in Cantwell Smith�s terms, how would
not the super-sunflower � what�s his other example???
rabbit and coyote � hmmm, there�s no representation there
sailor and the island�
me and the table
register without representations of some description�