Installed: 4 Jul 2014
Last updated: 4 Jul 2014
____________________________________________________________________________
This document is
http://www.cs.bham.ac.uk/research/projects/cogaff/misc/causation.html
A PDF version may be added later.
A partial index of discussion notes is in
http://www.cs.bham.ac.uk/research/projects/cogaff/misc/AREADME.html
____________________________________________________________________________
A: 2pm
Aaron Sloman
Computer Science, University of Birmingham
http://www.cs.bham.ac.uk/~axs
Welcome, introduce participants,
Introduce workshop:
Present some of the "standard" Philosophical background expanded by
Computational causation (e.g. causation in virtual machines composed of
multiple asynchronously interacting virtual machines)
Mathematical causation (e.g. moving a vertex of a planar triangle further from
the opposite side causes the area to be increased, adding three marbles to a box
containing five marbles causes the number of marbles to go up to eight,
changing the curvature of a line causes infinitely many distances between parts
of the line to change.
B: 2:25-2:35
Questions discussion and clarification, mainly for the benefit of
non-philosophers present.
Objections/counter proposals postponed until after the following talks.
D: Approx 2:35
Andrea Raimondi,
Philosophy University of Nottingham
http://about.me/Raimondiand
Functional Dispositionalism
Abstract:
I work specifically on causal dispositionalism and functions, but will try to
explain the metaphysics of causal powers, as background.
1. Dispositional Properties
According to dispositionalism, causation is produced by the exercise of
causal powers. The reason why dispositional properties are often
called powers is to highlight their causal role.
Dispositions inhere in the object of dispositional ascription. The causal
in causal powers is the fact that their exercise is productive of the
behaviour of things.
Dispositions are not mere potentialities. Their manifestations do not
depend on the presence of an active stimulus. To say that
something is soluble is not to think that the property is instantiated
only when the dissolving occurs (Mumford 1998).
Dispositions are active, but they produce the effect by composing
together with other dispositions. Causation as mutual
manifestation partnership explains the effect as the change in
powers when they team up. This starts a causal process that is
diachronical, with which we can interfere and that have as effect a
modification of the causal scenario.
2. Causal production
The exercise of dispositions represents the causal production. For
dispositionalism individual has their own causal powers and any
effects exhibited are the manifestation of those powers. But
objects possess dispositional properties even if those are not
manifested. To say that dispositional property is a property that can
exist unmanifested means that it points to objects' hidden
potentials.
A dispositional theory of causation sees discovery in the sciences as the
exploitation of causal powers and achievements in technology as
the ability to use objects' hidden dispositional properties to build
systems specifically designed to support a constrained set of
possibility.
The role "powers" remark is that dispositional properties are actual
causal properties that support object's potentialities. Causation as
the passing around of powers depicts hidden potential as a
disposition for the instantiation of another possible causal
property. That which is soluble has the potential to become
dissolved (Mumford 2009).
Causation explains the activity of individual objects in virtue of their
dispositional properties. Most of the complex biological
interactions that dispositionalism claims to explain involve causal
interaction.
a) Not all the causal interactions are also functional. Then how
a functional account of dispositions looks like?
3. Dispositionality as selection function
The idea of selection function is meant to explain that when powers
are exercised a subset of realm of possibilities is selected, favouring
some among others. Change in dispositional properties is explained
in terms of selection from a range of natural possibility
(Mumford&Anjum 2011).
Powers exercise provides the object with abilities to interact causally
with the world, linking set of possible changes. The modal space
available to the object is marked by the set of possible causal link
between natural possibilities (Sloman 1996).
Dispositional ascription allows causal explanation of behaviour of
natural things. Since dispositionality selects natural possibility, not
logical one, when powers are exercised we often know what to
expect. But dispositionality acts between pure contingency and
pure necessity.
A disposition only tends toward its manifestation, but not necessitate
the effect. We make sense of ascribing a dispositional property to
an object even without necessitating the manifestation.
3.1 Compositional pluralism and directenss
The same disposition can manifest differently in respect to the
presence or absence of other powers at work. Compositional
pluralism is the claim that there are pluralities of way in which
powers can compose to produce an effect. When they team up
they alter each other contribution to the causal outcome
(Mumford&Anjum 2011).
That a vase is fragile means there are possible ways in which the vase
can be damaged: it can be struck by a hammer, pushed on the
ground and smashed against the wall. There are also possible ways
in which the vase can tend to break: it can break in piece, or tiny
fragment, or burned to dust by a laser. Also a piece of metal could
be fragile. However for the fragility of metal to manifest requires
that other causal powers are in place. For instance, diminish the
temperature of the object beyond a certain threshold and it will
become as fragile as a vase.
Dispositionality explains objects' modal facts that are effects of the
powers control over potentialities. Powers act as a whole, at
different level of reality. A kiss may cause embarrassing, as well as
energy transfer between bodies. A cube of sugar can dispose
toward dissolving in water, as well as sweetening the solution.
b) Which one of this directness can be considered functional?
3.2 Getting functional
Functional Dispositionalism assumes that things in nature generate,
maintain and develop their own modal space and powers are
causal in that they constrain possible change as causation occurs.
Objects' dispositions depends on what natural possibilities are at hand
and which outcome is brought about depends of which functional
role can be instantiated. It is not obvious what it is for a disposition
to be a function.
A common assumption of dispositional ascription and functional
ascription is that both endorse a function/accident distinction.
Within this distinction the sweeten solution is an accidental effect
of the functional ascription of the solubility (Wright 1976).
Dispositional and functional are ascription that are always relative to a
choice of explanandum (Wright 1973).
c) Then, Is the claim that dispositional ascriptions are functional
ascriptions informative?
When we ascribe a disposition we are giving a functional
characterization of the property, according to what potentiality
support in different causal context. For something to be elastic
means to have a dispositional properties that cause its stretch
when pulled. We functionally characterize this disposition in terms
of elasticity.
Dispositional ascriptions are explanatory when they indicate what
potency is present in the object of ascription. Functional
characterization of this property does not exclude that same
disposition can manifests other effects accidentally.
Dispositionality as selection function causes possible changes in
powers composition. To ascribe a disposition is to say something
about what the objects can do, its modal space, not how it organize
it. To say that A and B mutually manifest C is not to say how A and
B compose C.
If an object must exhibit a function it is because an object has the
ability to do that. And to have ability is to have a disposition.
Nonetheless a dispositional ascription just specifies that a causal
contribution have a functional role, not how the contribution is
integrated when other powers at work realize the function.
Dispositionalist claim is informative in that suggest that dispositions are
particular kind of functions (Mumford 1998).
4. Dispositionality and Complexity
d) The realization of a function requires more than dispositional
ascription.
The modality of the system is represented as a selection function that
acts on various dispositions whose manifestations work toward a
common end-state.
It is because disposition is a particular kind of function that we can
admit the production of non-functional changes. For instance, the
appearance of an adaptive trait in an organism is strongly coupled
with the manifestation of non-adaptive trait. However those traits
might reveal themselves functional during organism development
(Nagel 1979).
For powers to instantiate a function they must be organized
appropriately. Is only when causal powers of many components are
brought together in the right arrangement that hidden
potentialities are supported, e.g. a bridge.
Not all causal inputs are for a selection function can be considered
fixed. When we pull ends of a climbing rope apart that could be
different pathway to end-state. There could be knots forming that
affect how the rope can be pulled. Its wetness can affect how the
rope curls.
A wire disposes to conduct electricity. But cooling down a piece of
metal and same dispositions manifest superconductivity, together
with the ability to store energy. With appropriate causal factor it
can dispose toward making light.
Where processes develops different constrain of possibility might
emerge and affect the paths to the end-state. Those constrain are
better understood as changes in complexity, affecting the way
dispositions can compose together.
In complex system functions always involve dynamically varying causal
inputs that are selected toward a casual output, each of those
modulates each other effect.
The ways component are organized have effects upon the selection
function, causing a change in complexity, both compositional and
dynamical. Functions in systems are involved with possible
interactions that change possibility between parts.
We can join two wires in parallel and their voltage is constrained to be
the same. When in series, the same happen in current. In both
cases, the disposition ascribed to the system could not be reduce
to the sum of the powers of the components.
Two wires coupled in a diode compose a complex object whose
functional characterization cannot be reduced to the sum of
functional characterization of component properties.
Functional dispositionalism can accept physicalism without imply
reductionism since powers act at different levels of complexity.
4.1 Adaptive Emergence
e) Powers exercise regulates the selection of naturally possible
outcome in respect to interference in causal ecosystem in
which the system behaves.
Functional systems are responsive to causal ecosystem in which they
behave. A functional account of dispositions can help investigate
emergent abilities of biological systems in terms of adaptive
response to environmental influence.
Organization only disposes toward the end-state since systems
behaviour is brought about by different dispositions working
together. Dispositional ascription instantiate more than one
functional role concurrently.
The adapting feature of powers composition can represent a step
toward the explanation of emergent change in complexity: evolved
complexity (Mitchell 2003). This complex feature of biological
systems can be interpret as incremental constrain over the modal
space generated by causal development.
References
Mumford, S. & Anjum, R. L., (2011). Getting Causes from Powers. OUP Oxford.
Mumford, S., (1998). Dispositions. Oxford University Press.
--- (2009). Passing Powers Around. The Monist 92 (1):94-111.
Mitchell, S., (2003). Biological Complexity and Integrative Pluralism. Cambridge Univ Pr.
Nagel, E., (1979). Teleology Revisited and Other Essays in the Philosophy and
History of Science. Columbia University Press.
Sloman, A., (1996) Actual possibilities. In Principles
of Knowledge Representation
and Reasoning: Proceedings of the Fifth International Conference (KR'96).
Morgan Kaufmann Publishers.
http://www.cs.bham.ac.uk/research/cogaff/96-99.html#15
Wright, L., (1973). Functions. Philosophical Review 82 (2):139-168.
---(1976). Teleological Explanations: An Etiological Analysis of Goals and
Functions. University of California Press.
followed by 5 minute break.
E: Approx 3:50
Comments from Al (and presentation of some of your own ideas if you wish)
How long would you like?
F: Comments by Alex on discussion so far
G: General discussion
END: approx 5pm, though if we over-run I expect the room will remain available.
Maintained by
Aaron Sloman
School of Computer Science
The University of Birmingham