[Martin Taylor 940902 11:50]
Avery.Andrews 940902.1216
(Martin Taylor 940901 12:40)
>Absolutely. See Bill Powers' posting (Bill Powers (940829.1700 MDT))
>with most of which I am in complete agreement, specifically with:
>>There is no entity in the
>>environment corresponding to the perceptual signal, nor need there be.
I don't quite agree with this - I see all `entities' as results of
analytical decisions that people make for specific purposes - `valid'
entities resulting when the analysis suits the purpose. So that, for
some purposes, its useful to think of floppy disks as containing
entities such as papers, data files, etc., whereas for others it isn't.
I won't try to put words in Bill's mouth, but I don't think there is really
much disagreement among us. It looks more like a confusion of how the
same set of ideas is worded. Let me try a dlightly different slant.
Assumption of the existence and consistency of reality:
We assume that there is a reality out there, and that whatever is there
has some consistent ways of working. At a slightly lower level of
confidence, we assume that these "consistent ways of working" don't
change much over lengths of time long compared with the life of a star.
They are sometimes called "Laws of Nature," which we can never know.
Assumption of the unity of reality:
We assume that we can access the "real" world only by means of our
sensory apparatus, and that the machines we build to sense things
that our personal sensors don't detect are accessing the same real world
as the one our sensors detect.
Process of testing of reality:
Because of the existence and consistency of reality, our actions can affect
the real world in ways that can alter what we sense. They can also affect
reality in ways that we cannot sense. And reality can affect us in ways
that we cannot sense directly (examples: intense gamma radiation, strong
fluctuating magnetic fields) though we can sense their effects indirectly
(sickness, hallucination...). Testing reality consists of determining
which actions have what consistent effects on our senses.
Construction of perceived reality:
The effects of our actions on reality may be consistent only with respect
to complex functions of what we sense. If we actually construct such
functions internally, we can use the consistency of our actions so as
to stabilize the values of those functions. We call that "controlling
perception." We can do this only when the functions and the actions
have some consistency, which depends on the real existence of at least
a temporary consistency in the way the real world works. When the real
world provides no such consistency between the actions and the outputs
of the sensor functions, we cannot control the function outputs. Our
"perceived reality" does not long incorporate such functions, since
actions that affect them have no reliable value for the actor.
Functions of reality and CEVs:
Whatever reality there might be, it is possible to conceive at least
an uncountable infinity of possible relationships among (functions of)
the entities in it. Any one of these functions could correspond to a
function internal to a perceiving organism, and thus to a "perception."
In this sense, one could say either that there are at least an uncountable
infinity of potentially perceptible "entities" in the "real world" or that
there are none other than those actually perceived by an organism. Both
statements are equivalent. I chose to use CEV as representing each of
the uncountable infinity of possibilities.
Significant functions of reality (ESVs):
Among all the functions of reality, some affect an organism. This means
that the value of the function may influence whether the organism lives
or dies, whether its actions continue to affect its perceptions consistently,
and the like. Slightly more formally, the value of an ESV affects the
value of some function whose arguments are entities totally within the
organism. These arguments may well themselves be the values of functions
whose arguments are in part outside the organism. The value of an ESV
may, but need not, affect the sense organs.
Perceptible entities:
Some ESVs have values that are perceptible, or are correlated with functions
that are perceptible. This means that they are functions totally of
arguments that depend on functions whose values are sensed by the sense
organs. A perceptible environmental function is a new construct in this
discussion, and I propose to call it a PEV (perceptible environmental
variable). Some ESVs are PEVs, and all PEVs are ESVs. The set of all
PEVs is a proper subset of the set of all ESVs.
A PEV is defined by a function internal to the organism, all of whose
arguments depend eventually on the outputs of sense organs. We call this
function a "Perceptual Input Function" (PIF). There exists in the real world
an entity corresponding to this function, one of the uncountable infinity
of possible functions that relate then unknowable elements of the real world.
The entity corresponding to the PIF is real, and is not itself a perception.
It might be detected and measured by machines with sensors sensitive to
the same functions of the environment as are the sensors of the organism
that defines the PEV.
Controlled perceptions:
When there is sufficient consistency in the real world that actions can
have an effect reliable at least as to direction on the value of a PEV,
the corresponding perception can be controlled. A PEV corresponding to
a controlled perception has been called a "controlled CEV" or CCEV, even
though the PEV itself is not what is controlled. It might be better to
call it a controlled PEV or CPEV, inasmuch as "CEV" refers to ANY function
of environmental variables, perceptible or not. Any PEV might possibly
correspond to a controlled perception at any moment, whereas almost all
CEVs could not.
What's an entity?
(Personal opinion): entities are "out there" in the real world. A CPEV is
an entity, no matter how complex the defining function. A tree is an
entity, but so are concepts like "democracy." Anything for which there
is a perceptual signal is an entity, which expands the list to include
all PEVs. I think there is no great value in extending the list to
include all ESVs, because whatever an ESV might be, nobody knows about
it until someone has perceived it, at which point it has become a PEV
for that person. ESVs just "are." The underlying thought is that the
notion of an "entity" includes the idea that the entity can be conceptually
isolated and discussed. "Entity" itself is a perception at or above the
category level. The entities for an organism are the PEVs and CPEVs defined
by the organism's perceptual functions. They are real and "out there."
Bottom line (of the argument and of the posting):
CEV = all functions of elements of the "real" world > ESV = all functions
that affect an organism > PEV = all perceptible functions > CPEV = all
functions corresponding to controlled perceptions.
Martin