Course Handout - Supervenience and the Mind-Brain Debate in Mental Philosophy
Copyright Notice: This material was written and published in Wales by Derek J. Smith (Chartered Engineer). It forms part of a multifile e-learning resource, and subject only to acknowledging Derek J. Smith's rights under international copyright law to be identified as author may be freely downloaded and printed off in single complete copies solely for the purposes of private study and/or review. Commercial exploitation rights are reserved. The remote hyperlinks have been selected for the academic appropriacy of their contents; they were free of offensive and litigious content when selected, and will be periodically checked to have remained so. Copyright © 2010, High Tower Consultants Limited.
First published online 17:30 BST 26th October 2005, Copyright Derek J. Smith (Chartered Engineer). This version [HT.1 - transfer of copyright] dated 18:00 14th January 2010
1 - How to Use this Glossary
It has become fashionable of late to apply the philosophical concept of supervenience, itself barely 50 years old, to the far more ancient mind-brain debate. Unfortunately for many of the interested onlookers, the resulting terminology (that of causal determination and formal logic) is often decidedly obscure. The purpose of this glossary is therefore to open up the lexicon of supervenience for the benefit of the non-philosophical disciplines within cognitive science. Entries have been cross-indexed using bold type in such a way that if loaded into a semantic network they would produce a navigable "data dictionary" for the chosen subject area. There is no "right" sequence to this navigation - the art is to wander, wonder, and then wander some more [except, perhaps, for the core entries on supervenience, where the "see firstly" / "see now" pointers are deliberately prescriptive]. Note also that although the body of material is reasonably self-contained, it sometimes overlaps in terms of content with companion glossaries, accessible by the hyperlinks provided.
2 - The Glossary Entries
Causal Closure: [See firstly domains and properties.] Kim's (1993) term for statements of the following form: "Any physical event that has a cause at time t has a physical cause at t" (p280). The essence of explanations based on propositions such as this is that the answer lies wholly within the domain in question; the domain, in other words, is causally closed. Physicalism is accordingly the prime example of causal closure at work.
Causal Diagram: [See firstly multiple causation.] A sketch-map of the decision points and optional event pathways in a complex causal line relationship.
Causal Line: [See firstly macrocausation vs microcausation.] A causal line is Bertrand Russell's conception of "a temporal series of events so related that, given some of them, something can be inferred about the others whatever may be happening elsewhere" (Russell, 1948, p459). Causal lines will not in fact be linear where (and to the extent that) the relationship in question involves multiple causation.
Causal Rule: A specific, coherent, and fallacy-free explanation of one or more event-pairs in a causal line. An explanation for which there exists a body of empirical support, and for which no alternative construction has been offered, even after sustained critical consideration.
Causality: For psychologists, the word "causality" usually refers to the compelling phenomenal judgement (once certain conditions of contiguity have been satisfied) that one event has resulted in the occurrence of another. So powerful is this judgement, indeed, that it will persist even when the experiencer knows that no cause and effect relationship exists, in the same way that visual illusions persist even once their illusory nature has been noted. The problem may be illustrated by considering the "stage punch", where actors feign fisticuffs without actually getting hurt. Actors throwing a blow, for example, deliberately swing an inch or so short, while actors being "hit" co-operate in the illusion by jerking the appropriate part of their body away at just the right time, and by crying out in pain. Carefully synchronised sound effects can be added as appropriate to intensify the illusion. Many theatrical special effects, conjuring tricks, and perceptual illusions work in similar ways. The effective variables were discussed more than half a century ago by the Belgian psychologist Albert Michotte (e.g. Michotte, 1946, 1963), and some vivid demonstrations are nowadays available in online simulation [try the keywords Michotte and causality]. Because causality is accordingly a "reserved word" in the psychological lexicon, it should be carefully distinguished from causation, the philosophical study of cause and effect. We may not enforce this distinction overzealously, however, because both words have been used in the cause and effect domain. Indeed, particular caution is needed when reading Kant's Kritik (or derivative works), because where Kant originally wrote Kausalität it has been brought across into English as "causality" when "causation" would have been more consistent (e.g. Kant, 1781/1787, p217).
Causation: "The action of causing; production of an effect" (O.E.D.). One of the fundamental principles of science is that there is regularity, reliability, and order in the natural world. Scientists then use the resulting predictability to control chemical reactions, calculate the positions of the planets many hundreds of years in advance, and generally bend the physical world to our will. Aristotle led the early theorising with his classification of four causes [tell me more] only to be conceptually superceded in the opening decade of the 17th century by Francis Bacon's new scientific method. The philosopher David Hume then provided us with the first integrated theory of causation in the mid-18th century. This theory is now commonly referred to as Humean causation, and includes the compulsory warning that the sense of causation which comes with detecting a cum hoc correlation between two types of event is often deceptive and counterproductive. And the reason it all matters, of course, is that "causation forms an integral part of evaluating explanations" (Park, 2003). [See now multiple causation and (for the real complications) supervenience.]
Cause: [See firstly causal line.] "That which produces an effect" (O.E.D.). The antecedent event in an event-pair for which a causal rule is being proposed; that which brought about the effect. Note, however, that with multiple causation there may be several contributory causes, all antecedent, and all converging (but not necessarily equally) onto the same effect.
Cause and Effect: See causation.
Contributory Causation: [See firstly causation and multiple causation.] The apportionment of causal influence amongst two or more causes during multiple causation. [Now compare necessary cause and sufficient cause.]
Cum Hoc Correlation: [From Latin cum hoc ergo propter hoc = "with this therefore because of this".] A type of fallacious argument in which a simple coincidence is confused with causation. [For examples of this and other types of fallacious argument, see our e-glossary on "Rational Argument".]
Domains and Properties: [See firstly supervenience and causation.] A domain is the set of entities currently under consideration in the framing of an explanation. A property is anything which can be claimed descriptively true of an event, for example, that Brutus' stabbing of Caesar occurred in Rome (Kim, 1993).
Dualism: See dualisms and monisms.
Dualisms and Monisms: [See firstly mind-brain debate.] A dualism is a "two-truth" theoretical position in the mind-brain debate, that is to say, one which claims that the laws of the mind and the laws of the brain are fundamentally irreconcilable. Monisms, on the other hand, are "one-truth" positions, claiming either (a) that the laws of the mind are the only real truth (in which case your monism is an idealism), or (b) that the laws of your brain are the only real truth (in which case your monism is a physicalism), or (c) that we are really not too sure (in which case you are probably going to be a monist one day, but are either an epiphenomenalist, identity theorist, or emergentist, for the time being). William James brought the dualism debate centre stage by referring very disparagingly to "mind stuff" theory, which he characterised as theories that mental states "are composite in structure, made up of smaller states conjoined" (James, 1890, pI.145). More recently, Velmans (2005) has referred to all the two-stuff theories as "substance dualism" because they are constantly pitting "material stuff" against "soul or spirit" stuff.
Effect: [See firstly causal line.] "Something accomplished, caused, or produced; a result, consequence" (O.E.D.). The "invariable consequent" (Mill, 1886, p213) event in an event-pair for which a causal rule is being proposed. That which has been caused. An event which is strongly supervenient upon an antecedent event (or events).
Elementism: Wilhelm Wundt's reductionist theory of perception, so called because it viewed perception as a synthesis of a number of individually less sophisticated sensory systems. Also, loosely, all Associationist theories of cognition, to the extent that they fragment knowledge into an agglomeration of sememes [glossary], propositions [glossary], images, and other primitive elements.
Emergentism: [See firstly mind-brain debate.] One of the three main not-quite-sure-yet philosophical positions on the mind-brain debate (the others being epiphenomenalism and identity theory). Like the Gestaltists before them, Emergentists rely on the concept of "emergent properties" to allow a whole which is greater than the sum of its parts.
Epiphenomenon: [See firstly mind-brain debate.] "Something that appears in addition" (O.E.D.). A by-product or accidental spin-off, but one which may nevertheless turn out to have a value in its own right. "A secondary phenomenon that results from and accompanies another" (The Free Dictionary). [See now epiphenomenalism.]
Epiphenomenalism: [See firstly mind-brain debate and epiphenomenon.] One of the three main not-quite-sure-yet philosophical positions on the mind-brain debate (the others being emergentism and identity theory). Epiphenomenalists subscribe to the notion that "mental events are caused by physical events but have no causal effects themselves" (Gray, 1987, p462). Alternatively, "the classical form of epiphenomenalism [denies] that mental-to-physical causal action ever takes place .....] Mental phenomena are totally causally inert" (Kim, 1993, p104; emphasis added). In fact, Kim doubts that epiphenomenalism is a valid position in the first place, seeing it as fatally flawed logically. He follows Lachs (1963) in arguing that the very fact we are able to discuss events implies that they have caused that at least! Nevertheless, the term epiphenomenon is regularly encountered in the mind-brain debate, because it allows the mind to be treated as an emergent property of this or that underlying neural activity. On balance, however, many authors regard identity theory as superior in that it does not rule out mental-to-physical causation quite so high-handedly. One of Kim's (1993) observations puts the central issue very succinctly, thus: "Given that any physical event has a physical cause, how is a mental cause also possible?" (p281; italics original). He calls this the problem of "causal-explanatory exclusion".
Event: In everyday usage, "anything that happens or is contemplated as happening; an incident, occurrence [.....]; that which proceeds from the operation of a cause; a consequence, result" (O.E.D.). In philosophy and science the same, but (1) as part of a more general search for the fundamental laws of causation in nature, (2) inspired by the belief that this search will proceed more profitably if guided by the scientific method (rather than, say, by meditation or oracular consultation), and (3) with a constant battle against "events" turning out upon closer inspection to consist of lesser events. Kim (1993, p4) reduces the philosophical issue to the following question: "In what relation must a pair of events stand to a law [i.e. of causation] if the law is to 'subsume' the events?". Kim also points to an interesting hole in the scientific method, namely that it is deceptively easy to overfocus on events to the exclusion of the matrix of states - the uneventful times - within which the events take place. [See now macrocausation vs microcausation.]
Event-Pair: [See firstly event.] The causes and effects in a simple instance of linear causation (or, in an instance of multiple causation, each one of the causes converging on the same effect).
Explanation: "That which explains, makes clear, accounts for [..... especially] with a view to adjust a misunderstanding and reconcile differences" (O.E.D.). "An argument to the effect that the phenomenon to be explained [.....] was to be expected by virtue of certain explanatory facts" (Hempel, 1965, p336). But more besides, because you owe it to your audience (and yourself) to set the core explanation in as broad a context as possible. So, alternatively, a collection of proposed causal rules which, taken together, state the "particular 'go'" of a natural phenomenon, or a statement not just of how something works, but of why it makes broad sense for it to work that way given what else we know about the world. This latter quality is akin to what Kant (1781/1787) called the "systematic unity" of the best scientific understanding: our cognitions, he warns, "must not amount to a rhapsody; rather, they must amount to a system" (Op. cit., p755). Example: To experience for oneself what is involved in producing a thorough explanation, try this long-running poser: How much does smoke weigh, and why (after Kant, 1781/1787, p255).
Explanatory Gap: [See firstly explanation and reductionism.] The explanatory gap is the conceptually grey area between the microdata of neurophysiology and the macrodata of philosophy and psychology. It became a popular topic of scientific discussion in the mind-brain debate following Levine's (1983) focus paper on the subject, and the central problem seems to be that our minds cannot grasp how something as intangible as a mind might be reduced to individually straightforward electrochemical events. Our minds do not feel reducible, so we convince ourselves that that cannot be how they work. For our own part, we suggested in Smith (1998) that the explanatory gap could profitably be regarded as a "compiler gap". Our point was that computer engineers in the 1950s had unwittingly solved some major neurophilosophical problems. They had built logic circuitry capable of executing microinstructions, only to discover that this low-level language was so tortuous to work with that they needed a language of macroinstructions to convert it into usable chunks (it was, for example, far more efficient to work with the single high-level instruction <ADD> than with the hundreds of low-level machine instructions it translated into). Compilers were the systems software products which carried out the necessary translation, and by the early 1960s were allowing applications programmers the luxury of doing their technical problem solving in near-natural language. Cognitive scientists have yet to exhaust the compiler metaphor as a means of correlating the "source code" stream of ideation with the "object code" stream of the underlying cellular and subcellular biochemistry. This could well turn out to be an unfortunate oversight because source code supervenes precisely onto object code, and object code supervenes precisely onto activity in the underlying circuitry!
Global Supervenience: See supervenience, weak, strong, and global.
Humean Causation: This is the name given to the theory of causation advanced by the philosopher David Hume in Book I of his Treatise, which proposed the entirely commonsense scheme of "contiguity, succession, and necessary connection" (Hume 1739a, p77) (in other words, causes have to precede their effects, and be close enough to them in time and space for a possible causal line to be detected and agreed). Hume himself pointed to a number of critical weaknesses in the commonsense scheme, such as the near certainty of transitive causation and multiple causation, and the risk of falling for a cum hoc correlation fallacy. For these and other reasons, commonsense explanations of causation often do more harm than good to the progress of science, and Hume's position was duly developed over the ensuing decades by the likes of Immanuel Kant's Kritik and John Stuart Mill's System of Logic. Kant saw cause and effect as part of the broader acquiring of understanding, whilst Mill saw it as that all-important grasp of "reasonable anticipation of future facts" (Mill, 1886, p212) characteristic of the good scientist. More recently, Hempel (1965) has incorporated probability theory into the equation by allowing for laws to be based on the probable contribution of individual causes rather than absolute determination.
Idealism: One of the two possible monist positions in the mind-brain debate (the other being physicalism). Specifically, the notion that the laws of the mind will, once they have been finally and fully established, be able to explain not just the workings of the mind, but those of the brain as well.
Identity Theory: [More precisely, mind-brain identity theory or psychophysical identity theory.] [See firstly mind-brain debate.] One of the three main not-quite-sure-yet philosophical positions on the mind-brain debate (the others being emergentism and identity theory). Identity theorists subscribe to the notion that "mental states are identical to brain states and that "a given mental state will be fully accounted for if and when one has accounted for the corresponding brain state" (Gray, 1987, p461). Much of the modern interest in the subject can be traced to a paper by Place (1956) which asked how we could know whether two sets of observations were of the same event. Was a cloud, Place asked, the same thing as the mass of tiny particles making it up? Clearly it all depended on how you happened to be looking at it at the time. His point was fundamentally "that an acceptance of inner processes does not entail dualism" (p43). In a "second look" at the issue, Gray (1987) advises that we do not yet know enough about "the conditions for consciousness" (p480) to test any of the alternative explanations. It may well be, he argues, that the machinery out of which a cognitive skill is delivered is genuinely "unimportant", and that what is critical instead is "the nature of the skill itself" (p482).
Identity Theory, Kim's Position On: [See firstly identity theory.] Kim generally takes a "modified reductionist" view of the mind-brain debate, and usually casts his vote with the identity theorists. Here is one of his arguments in full: "Consider saying that there are in this glass two distinct substances, H2O and water; that is, consider saying that water and H2O co-occur everywhere as a matter of law but that they are distinct substances nonetheless. This would invite a host of unwanted and unnecessary puzzles: given that what is in the glass weighs a total of ten ounces, how much of the weight is to be attributed to the water and how much to the H2O? By dropping a lighted match in the glass, I extinguish it. What caused it? Was it the water or the H2O? Were they each only a partial cause, or was the extinguishing of the match overdetermined? The identification of the water with the H2O puts all these questions to rest" (p281). To see Kim's substantive position, simply re-read this quotation, substituting "skull" for "glass", "brain" for "H2O", "mind" for "water", "problem" for "match", and "solve" for "extinguish".
Macrocausation vs Microcausation: [See firstly causal line, event, and explanation.] The problem with offering scientific explanations is that they are inevitably phrased by particular authorities for particular audiences. One's interpretation of an observed phenomenon therefore tends to be coloured by an equally particular presumption of the relevant event sequence. It is therefore important to remember that Russell's notion of the causal line is only a convenient approximation, because the "events" in question invariably turn out upon closer inspection to consist of lesser events, and they of even lesser ones, and so on. Macroevents, in other words, consist of microevents, and macrocausation cannot finally be established without establishing a chain of microcausation. Kim (1993) explains why we need to worry: the difference between the two approaches, he points out, lies in the fact that macrocausal relationships are supervenient upon microcausal ones. He gives the example of a sudden pain in the thumb followed by a sharp withdrawal reflex. The microstimuli on this occasion follow "the usual physiological causal path" (p103), but the experience of the accompanying pain needs to be considered separately. In the withdrawal reflex, therefore, pain is an effect, not a cause, despite subjective evidence to the contrary.
Materialism: Same as physicalism. "An absurdity" (Eccles, 1987, p293).
Mereological Supervenience: [See firstly domains and properties and supervenience.] "The doctrine that the character of a whole is supervenient upon the properties and relationships holding for its parts" (Kim, 1993, p113). Kim goes on to point out (a) that this sort of supervenience will always have to cross two domains by definition, "one domain consisting of wholes, and another consisting of their parts" (Ibid.), and (b) that we still have a lot of work to do in deciphering how this sort of supervenience works in practice [although the possibility of biological semaphores and busy pins offers some grounds for early optimism here.]
Mind-Brain Debate: This is the formal name for humankind's age-old quest to understand how that which we experience at first hand as the workings of our mind (i.e. our perceptions, emotions, memories, insights, etc.) might conceivably be supported by the "two fistfuls of porridge" (Taylor, 1991) which is our brain. The fundamental problems are (a) that we do not have experiential access to most of what goes on in our mind (to borrow one useful current phrase, most of that lower activity is "transparent" to our introspections), (b) that even when introspection is successful it is by definition impossible for it to be independently validated, and (c) that we are not yet good enough engineers to fathom out the brain's operating principles [that which James Clerk Maxwell in the 19th century and Kenneth Craik in the 20th liked to call "the 'go' of it" (Sherwood, 1966)]. Or to put it another way, there is a lot of mind which never experiences anything, but just goes happily about its work. As a result, there have always been fundamentally different competing views on the mind-brain relationship, as introduced by the separate entry for dualisms and monisms. [See now supervenience.]
Mindness: This is Llinás' (1987) notion of a "high-level awareness, including self-awareness" (p356) which allows "complex goal-directed interactions between a living organism and its environment" (p339). [See now how Llinás uses mindness in defending his own brand of physicalism.]
Monism: See dualisms and monisms.
Multiple Causation: [See firstly cause, effect, and event.] A non-linear configuration of related events, such that any of several causes is capable of making an effect more (or indeed less) likely to occur. A "composition of causes" (Mill, 1886, p212). Whenever multiple causation is suspected, it brings with it the need to classify each of the several causes as either a necessary cause or a sufficient cause. Proposed causal configurations can often be made more apparent if their key points are expressed graphically in causal diagrams.
Multiple Domains: See mereological supervenience.
Necessary Cause: An instance of multiple causation in which one of the causes does not just contribute towards the occurrence of the effect, but is a sine qua non of its occurrence. Or to put it another way "causes are often distinguished into two types: necessary and sufficient. If x is a necessary cause of y, then y will only occur if preceded by x. In this case the presence of x does not ensure that y will occur, but the presence of y ensures that x must have occurred. On the other hand, sufficient causes guarantee the effect. So if x is a sufficient cause of y, the presence of x guarantees y. However, other events may also cause y, and thus y's presence does not ensure the presence of x" (Wikipedia). [Compare sufficient cause.]
Physicalism: One of the two possible monist positions in the mind-brain debate (the other being idealism). Specifically, the notion that the laws of the brain will, once they have been finally and fully established, be able to explain not just the workings of the brain, but those of the mind as well. The position is interpreted by Llinás (1987) using his notion of "mindness", as follows: "I for one, as a monist, consider 'mindness' to be but one of several global physiological computational states that the brain can generate. An example of another global physiological state, in which 'mindness' is not apparent, is that known as 'being asleep' and yet another is known as dreaming" (p339).
Ports, Pins, and Drivers: It is in the nature of modern electronic devices that their substantive logic circuitry is chip-mounted onto printed circuit boards, the circuit boards slotted into a larger chassis, and the chassis equipped with the necessary communications sockets (or "ports"). A system of internal wiring (the "loom", or "bus") then connects separate mounting boards to each other and/or to the communications sockets as appropriate. The connections between the electronics components and the circuit board, and between the circuit board and the wiring loom, are called "pins" and "pinouts", respectively. Software modules written specifically to initiate output to, or interpret input from, pinouts, are known generically as "device drivers" (or just "drivers" for short). [See now semaphores and busy pins.]
Reductionism: [See firstly explanatory gap.] A philosophical doctrine predicated upon the assertion that complex sociocultural and psychological phenomena can ultimately be explained in terms of underlying chemical or physiological processes [but far from universally supported].
Scientific Method: [See firstly causation.] The use of inductive reasoning in the pursuit of broader and more reliable knowledge. The roots of the scientific method go back to Sir Francis Bacon, and are based upon the repeated, published, and peer-reviewed empirical testing of hypotheses.
Semaphores and Busy Pins: [See firstly ports, pins, and drivers.] A semaphore is a one-bit short-term memory store, addressable from the inward direction by the local processor, and directly connected in the outward direction to a communications pin of some sort. The value of this contrivance of electronic engineering is that it allows the local Control Unit to monitor the processing state of a remote module with which it is for the time being associated. Semaphores are typically used to synchronise the execution of logically related software in distributed processing systems. This means getting the semaphore (in essence, only a single binary flip-flop) and the pin (a physical connector in a larger modular circuit) to work together, and the classic solution for half a century has been for the processor in question (a) to monitor the processing state of an associated module by wiring the latter to a pin it can itself directly access, (b) to monitor said pin, and (c) to schedule its own processing according to what that monitoring reveals. The signal pins are known generically within electronics as "busy pins", and the timing instructions are of the nature <WAIT ON BUSY PIN HIGH/LOW>. The semaphores - one per unit being monitored - then simply echo that pin status into the software, where it can be tested (in other words, this clever piece of engineering makes the physical mental, and vice versa - see endnote). To maximise processing speed it is usual to maintain the semaphores at bit level rather than at byte level, and to provide them as a class of purpose-built CPU registers called "status registers". Heidenstrom (1998) shows several nice tables relating bit values within the status registers to the numbered pins on the communication ports, if interested.
Note: Semaphores and busy pins have a major but as yet unrecognised metaphorical relevance to the mind-brain debate. This is because they have been having mereological supervenience designed into them since they were first invented in the 1940s, and must now be rated as tried and tested mechanisms for interfacing process with underlying processor. For whatever reason, however, there has been little explicit search for the biological equivalent of semaphores and busy pins, to the detriment of the central debate. [See now supervenience in modular processing hierarchies.]
State: [See firstly event.] "Static things" (Kim, 1993, p33). The quiet times between events, and a potential problem to theories of causation. Kim, however, plays down the differences between events and states. He sees the former as "a losing or acquiring" some important property and the latter as simply "having" it at a particular moment in time.
Strong Sufficiency: See sufficient cause. [Do not confuse strong sufficiency with strong supervenience.]
Strong Supervenience: See supervenience, weak, strong, and global.
Sufficient Cause: [See firstly multiple causation.] The notion that event A is capable of triggering effect Z in isolation, but that it would find it easier to do so if at least one other contributory cause was simultaneously active. Where the need for contribution is in fact relatively small, then event A may be described as "strongly" sufficient. [Compare necessary cause.]
Supervene, To: [Latin supervenire = "to come/occur above".] "..... to follow closely upon some other occurrence or condition" (O.E.D.). This word is rarely (if ever) used in everyday English, but its formal meaning is worth noting carefully because it is the source of the much trickier philosophical term "supervenience". [See now supervenience and causation.]
Supervenience and Causation: [See firstly causation and supervene, to.] Philosophically speaking, supervenience involves a supervening of elements of two sets of properties from a single domain in accordance with a causal rule. Note that this usage of supervenience immediately goes further than the standard English usage: specifically, for one event to "supervene" upon another in a philosophical sense, it has to do more than just follow that earlier event, rather it has to occur because of it. For a thorough history of the derivation of the philosophical usage, leading into a detailed discussion of the dozen or so different facets of supervenience across various branches of philosophy, see the (18-page!) entry in the Stanford Encyclopedia of Philosophy [click here to be transferred] or work through the essays in Kim (1993).The common denominator, however, is that supervenience indicates consequence rather than mere sequence, and its usage in this sense dates from the 1940s (Hare, 1984). Defined formally, philosophical supervenience involves the cross-relating of properties between "two sets of properties over a single domain of individuals" (Kim, 1993, xi; italics original) and signifies "a metaphysical and/or conceptual determination-relation" (Horgan, 1993, p555). [See now supervenience, weak, strong, or global.]
Supervenience for Multiple Domains: See mereological supervenience.
Supervenience in Mental Philosophy: [See firstly supervenience, weak, strong, or global.] Nowhere is there a greater need to master the language of supervenience than in modern mental philosophy, where the challenge is to identify which of the two properties (i.e. mind events and brain events) actually supervenes in the causal line of biological cognition. Davidson (1970) opened this phase of the debate with the following general declaration: "Mental characteristics are in some sense dependent, or supervenient, on physical characteristics [although] dependence or supervenience of this kind does not entail reducibility through law or definition" (p88). Supervenience soon became a major analytical tool in all subsequent discussions of reductionism and the explanatory gap, and inspired works by the University of Sydney's John Bacon (e.g. 1986), Brown University's Jaegwon Kim (e.g. Kim, 1993), the University of Arizona's Terence Horgan (e.g. Horgan, 1993), and Princeton's David Lewis (e.g. Lewis, 1991). Kim's (1993) studied dismissal of epiphenomenalism [see that entry] offers a good example of how the concept can inform the mind-brain debate when it has to, unfortunately there is still too much uncertainty about whether brain events follow mental ones, or vice versa, to make the long-awaited major breakthrough. [See now supervenience in cognitive processing hierarchies.]
Supervenience in Modular Processing Hierarchies: [See firstly supervenience in mental philosophy.] And so to the real point of this glossary, which is that it is impossible to apply the notion of supervenience to the mind-brain debate without immediately coming up against the intricacies of modular processing hierarchies. The critical point is that neither the "mind" nor the "brain" is grainless - they both have internal structure (that of the brain, indeed, having been documented in exquisite detail over the last two centuries or so). They therefore suffer the problems of macrocausation vs microcausation not just once, but repeatedly, once for each identifiable module. In other words, when Rumelhart and McClelland (1985) popularised the notion of parallel distributed processing they were - in effect, if not intent - introducing new microlines of causal determination for each distributed processor, and then insisting on a superordinate macroline for the system as a whole! Again, any search for supervenience between the two sets of properties will run into difficulty when trying to unravel the all-important event sequences, because there will be simultaneously advancing, but different, event sequences in each module for each set of properties. Computers overcome these difficulties by resorting to semaphores and busy pins, plus an awful lot of very expensive "netware" to manage them.
Supervenience, Weak, Strong, or Global: [See firstly supervenience.] One of the most valuable subthemes in Kim's (1993) review of the subtypes of supervenience was its distinction between weak, strong, and global versions thereof. Supervenience, he argues, is not a black-or-white quality because natural phenomena are not black-or-white. Granted a few causal lines are relatively well mapped out, but most are horribly confounded by the problems of macrocausation vs microcausation. If we are to make any progress at all, therefore, we need to settle for less than perfect explanations, and allowing weak supervenience (i.e. supervenience where set A properties co-vary in terms of "discernibility" within a single world [see longer definition]) is one way to achieve this. Strong supervenience is then simply weak supervenience with a higher proportion of inter-world dependencies. "Global" supervenience, by contrast, is simply the non-specific requirement that "worlds that are alike in a certain way must also be alike in another way" (Bennett, 2004, p501). [See now supervenience in mental philosophy. For an example of one hundred percent strong supervenience, see our point about source code and object code in explanatory gap.]
Transitive Causation: [See firstly causal line.] This term derives originally from the works of Benedict de Spinoza in the 17th century, and denotes a causal line in which various intermediate stages are not necessarily noted by an observer. Park (2003) puts it this way: "Intuitively, we expect causation to be transitive. If x causes y and y causes z we expect x to cause z" (p3). [See now the discussion of microevents in macrocausation vs microcausation.]
Weak Supervenience: See supervenience, weak, strong, and global.
3 - References