RE: BHA: Emergence

["Marshall Feldman" <marsh@xxxxxxx>]

Hi all,

This is a long post, so be warned.

Mervyn wrote:


> There is no requirement that the emergent level does react back on the
> lower level, only that it can do - and life-forms certainly can. I'm not
> sure that emergence is not always diachronic as well as synchronic.

Neither am I sure, although I suspect you and I lean towards different sides
of the question. This certainly would be an interesting question to explore.

I'm tempted to say that synchronic emergence is a movement towards
increasing complexity. The logic behind this is that synchronic emergence
concerns levels in the world. As you rightly say, the properties and causal
powers of things at one level cannot be derived or predicted from those
things that comprise it. Their specific combination and relation to each
other, what RB calls structure, is what accounts for the causal powers at
the higher level. Since things at a higher level of combination include
things at a lower level, it would seem that emergence always involves
increasing complexity.

The problem with this is that quantum physics teaches us that this is not
necessarily so. Water molecules are, in some sense, less complex than oxygen
and hydrogen atoms, and they in turn are less complex than subnuclear
particles. However, I'm using "complexity" here in a way that trades off
"human understanding." High school students learn about water molecules
because, among other things, it's easy to construct a physical model with
one large blue ball linked to two smaller red balls which, if memory serves
me, are at a 120 degree angle to each other. On the other hand, try
explaining to a high school student that an electron is really mass-energy
distributed in space according to a 3-D probability density function!

So this raises another question. Is complexity inherently anthropocentric?
Logically, since higher-level beings must contain the lower-level things
that comprise them, higher-level things are more complex almost by
definition. However, precisely because of emergence, the properties of
higher-level things may (partly) result from those of lower-level things,
but these properties themselves may be less complex (in senses such as
taking on fewer different states, changing less frequently, having fewer
distinct causal powers, etc.)? In other words, can we (1) fashion a
non-anthropocentric notion of complexity and (2) say that complexity is
itself an emergent property so that we can separate the concept of emergence
from that of complexity?

Another concern I have is that so much of this discussion subtly extends
natural science, particularly physics, to other domains. We come to
understand the concept of emergence through examples like water. Once we
grant it legitimacy we then use it as a principle elsewhere. While accepting
the concept of emergence and allowing for it *possibly* to be at work
elsewhere, I wonder if in fact our notion of emergence needs modification in
other realms. This may account for why we're debating how to think of
emergence in relation to ideas and to society.

Even the term, "emergence," is perhaps loaded with a residue of reductionist
baggage it ought not carry. It implies that properties at one level *emerge*
from properties at a lower level. What if we just said that reality is
layered in different ways, one of which is that the world has "bigger"
things consisting of "smaller" things; these bigger and smaller things are
what we mean by "levels," and properties and causal powers at one level
cannot be predicted by or deduced from those at another level? This would
leave open questions of historical sequence, of what emerges from what, and
most generally of chickens and eggs. In the social sciences at least, both
Marxists and feminists have argued that individual personality emerges from
societal forms. Indeed, RB's observation that emergent things can react back
on the thing from which they emerge makes this a distinctly chicken-egg kind
of problem.

The reason why I still want to separate level and see emergence as a
synchronic relation between levels is to avoid the chickens and eggs.
(Sorry, it must be the holidays that are making me use this metaphor; it's
their causal powers emerging in my writing -- MF ;) ). Think of the TMSA. RB
introduces history here as a step forward from earlier "dialectical"
formulations such as Berger and Luckmann's. This move is justified, but I
worry when we start to conflate questions about what came first in order to
answer questions about the relationship between levels. I still think
there's a qualitative difference between on one hand the relationship
between capitalism as a social system and individual characteristics such as
personality, "rationality," etc. and, on the other hand, between capitalism
and feudalism, its predecessor.

Viewing the problem this way makes disentangling things much more complex,
but I think also more perspicacious and, ultimately, more warranted (i.e.,
more true). Take RB's picture of the TMSA (p. 77 in RR):


          -------------> Society --------------------->
                     ||               ^^
                     ||               ||
                     vv               ||
          ------------->Individuals ------------------>

We have to recognize that at any one point in time, both synchronic and
diachronic emergence are at play, BUT THEY ARE NOT FROM THE SAME THINGS. To
make this point, reproduce a portion of the picture:


                      Society -------O---------->
                                    ^^
                                    ||
                                    ||
                      Individuals -------------->

where O is the social reality we're trying to grasp. It emerges out of BOTH
society's history (diachronic) AND individuals present in that society
(synchronic -- both in terms of action and personality). Of course previous
society itself emerges out of individuals, but (as the downward arrows
indicate) those individuals themselves emerged out of previous society.
Hence, from the standpoint of O, previous societies mediate previous
individuals, and contemporary individuals mediate past societies.
Nonetheless, in the first instance a society emerges out of its history qua
society and the individuals that comprise it. The society's history (e.g.,
the inherited physical structure of human settlements, social institutions,
etc.) constrains, empowers, and problematizes the society today while in a
very different sense, the people who comprise that society have similar but
different effects. To take a concrete (pun intended) example, what will
happen in lower Manhattan next year will be due to the events of 9/11, the
physical layout of Manhattan, the socioeconomic institutions of capitalism
and the stock and real estate markets (assuming they're still here next
year), as well as due to individual decisions such as the ballots cast by
NYC voters, the decisions to move or stay in the city, or the policies
adopted by the new Mayor.

In this example, "emergence," for want of a better word, has three distinct
senses that we should keep separate: (1) history (the horizontal arrows --
e.g., 9/11), (2) deliberate action (the vertical arrows), and (3) unintended
consequences (the vertical arrows again, at least those pointing up). I've
always understood emergence to be only #3, but then perhaps we should
include the other two and assign special names for them. I propose:
historical, active, and inter-level emergence.



> Even
> in your synchronic case of water, I take it that there was a time when
> hydrogen and oxygen existed but no water, and before that various gases
> and particles, but no hydrogen or oxygen... DPF is clear that both
> synchronic and diachronic explanatory reductions of emergent phenomena
> are possible, but only after the emergent level has been understood in
> its own terms. This entails seeing that it could not have been predicted
> (induced or deduced) from the materials out of which it is formed, i.e.
> that is has genuinely novel, irreducible causal powers. In my 'society'
> example I was thinking, not of the society/person relation, but of the
> emergence historically of different types of society, as grasped e.g. by
> the concept 'mode of production'. Thus capitalism is diachronically
> emergent from feudalism, and certainly capable of reacting back on it
> (something still occurring today in relation to feudal remnants).


Can you fit the discussion of feudalism into the pictures above? I think the
only way to do this is to pose either a geographic separation or to think of
"society" as in fact being a social formation in which distinct modes of
production exist contemporaneously and conterminously, but at the level of
society rather than at the level of individuals. (It never ceases to amaze
me how much of this brings us back to Althusser.) I think what you're
talking about would then have to be yet a fourth level of emergence, to wit:

          -------------> Feudalism  - - - - - - - - - - ->
                     ^|               ^|
                     ||               ||                      } Society
(social formation)
                     |v               |v
         - - - - - - - > Capitalism --------------------->
                     ||               ^^
                     ||               ||
                     vv               ||
          ------------->Individuals ------------------>

Here the space-out lines, - - -, indicate ascendant and declining modes of
production.

If we do this, then I think it becomes clear that the feudalism capitalism
emerges from is not the same feudalism that capitalism reacts upon. They are
separated in time. On the other hand, at any point in time the social
formation is emergent from the specific combination of modes of production
as well as from the individuals who comprise it through their everyday lives
and through their intentional actions aimed at the societal level.

>
> Tobin:
>
> >> >I don't think one can use explanatory power
> >> >as a mark of ontological difference among ideas, because the
> explanations
> >> >concern referents.
>
> Mervyn:
>
> >> But you hold that ideas are not *the same* kind of thing as their
> >> referents; further, as Ruth said, belief systems (can) presuppose other
> >> belief systems.
>
> Marshall:
>
> >I'm not sure I get the point of this. Can you elaborate on how
> difference in
> >kind between ideas/referents or the presupposing of prior belief systems
> >pertains to the issue of using explanatory power to determine ontological
> >differences?
>
> Tobin says that explanatory power can't be used to indicate ontological
> difference among ideas *because* explanations concern referents. I take
> this to be Ruth's point about content. He seems to want to say that
> where there are emergent levels it is in the intransitive objects of
> ideas, not at the level of ideas themselves. My point to Tobin is that
> this seems to reduce ideas to their referents or content (the ontic
> fallacy, I think), and is at odds with his insistence on the social
> production of knowledge by means of knowledge via the transformation of
> pre-existing materials. While the production of knowledge occurs in the
> transitive or epistemological dimension, it can of course be
> referentially detached and studied as an intransitive object. In which
> case, it can be seen to involve the emergence of a new (conceptual)
> being (dpf 49) or theory out of the pre-existing materials of the lower
> level and possessing powers that are irreducible.

I think I see your point here, but again I think we need to look more
closely. First, let's forget about referents. I agree that they're besides
the point when we consider emergence at the level of ideas. (Although I'll
reintroduce them in a minute.) Second, I would grant you that the concept of
emergence works at the level of ideas alone, but I would hasten to add that
the synchronic/diachronic elements in the TMSA are also at play here. E.g.,
neoclassical economics emerged out of classical economics (and in opposition
to Marxian economics), AND the properties of neoclassical economics (e.g.,
law of supply and demand) emerge from the theory's building blocks (e.g.,
rational economic man [sic] and, at a lower level, a rationalist,
masculinist empiricism which deems deductive/analytic modes of thought
superior to others). Both kinds of emergence happened, but they're not the
same kind.

Now let's bring the referents back in. They too can determine thought. When
the Great Depression hit, Keynes went scurrying to explain what neoclassical
economics could not. Should we say that Keynesian economics emerged from the
Great Depression?

I think we're beginning to see the problem here. The notion of emergence can
mean many distinct things, and if we're careful there's nothing wrong with
using it these different ways. For the sake of clarity, I prefer to reserve
"emergence" to designate a NECESSARY relation, whereas many of the other
meanings discussed above denote either contingent or even accidental
relationships (by all accounts, the fact that the WTC collapsed was an
accident that nobody, not even the terrorists, planned -- this is not to say
the terrorists were not delighted at the accidental results -- at a
different level, that of the building's mechanics, the collapse was
necessary). I also prefer to distinguish emergence between levels from
historical, logical, and other forms of emergence. Thus, while the
occurrence of a specific configuration of things at one level may be
contingent, once we have this configuration there's a necessary relationship
to other levels. This is not to deny that the other relationships do not
exist, it's only to say that we'd gain more clarity if we clearly
distinguish between the different meanings of "emergence."

> >To me it seems the very nature of "explanatory power" puts ideas in a
> >separate category from most (other) ontological things. While
> along with H.
> >Marcuse, I do not want to abolish ideas from being part of the world, I
> >nonetheless think they are a special category and we need to be
> very careful
> >when we refer to their ontological status.
>
> Of course they are in a special or distinct (not separate, because they
> emerge from pre-existing materials) category - they are emergent, or at
> any rate the product of an emergent power of matter: mind.

Right, but everything in the universe is connected to everything else. I'm
not sure we gain much by stressing the mind-matter relationship when we
discuss the relation between ideas and referents or between ideas and their
antecedent ideas or between ideas and their presuppositions. The only time
the mind-matter relationship warrants our attention is when we really want
to say something about it in relation to the other things at issue.
Otherwise, it's just a diversion.

>
> Marshall:
>
> >Yeah, but I don't think anyone on this list buys into this kind of
> >irrealism. It's a straw man.
>
> Well, Marko has got the impression that they do, and this is what I'm
> trying to correct. If people keep intoning that knowledge is a social
> production (in the transitive dimension) without pointing out that it
> does *not* produce the objects of knowledge in the intransitive
> dimension, of which genuine knowledge is possible, it's not going to be
> corrected and we pretty inevitably come across as social
> constructionists in the above sense.

Yes, I can see this. But this whole issue is so delicate today that it poses
a problem. On the one hand, we do have the postmodernists and the whole
irrealist thread running through their discourse. Lots of this is just
nonsense or language games of a very superficial nature. On the other hand,
some of their concerns about language, the theory-laden nature of
observation, the role of scientific communities, etc. make considerable
sense and deserve to be taken seriously. The difficulty is that so much of
the latter, good stuff slides into the former, nonsense. CR has the unique
(to my knowledge) advantage that it can incorporate so much of the cultural
(Kuhn)/linguistic (Quine, Davidson) critique of positivism while still
insisting on a reality outside culture and language and a notion of truth
that is not relativist. These are not simple waters to negotiate, but emails
become excessively long (like this one) if we have to present the whole
argument every time we discuss anything.


	Marsh Feldman



     --- from list bhaskar@xxxxxxxxxxxxxxxxxxxxxxxxxx ---