Supervenience
from 1952 Hare's use in ethics; the modern technical sense is Davidson's and Kim's
If two cases are identical in every underlying respect, they cannot differ in the respects that depend on those — so any difference up here demands a difference down there.
- Active
- from 1952
- Where
- Oxford, later Michigan
- Status
- In use
- Tradition
- Analytic tools
What it is
One set of properties supervenes on another when there can be no difference in the first without a difference in the second. Fix everything at the lower level and the upper level is fixed too.
The argumentative use runs backwards. If you claim two cases differ in some supervening respect, you have committed yourself to a difference in the base — and if you cannot produce one, your claim is in trouble.
How it works
Establish that the upper properties supervene on the lower ones. Take two cases claimed to differ upstairs. Demand the corresponding difference downstairs. If none can be produced, either the claimed difference is not real or the supervenience claim is wrong.
The force comes from the demand being unavoidable. It is not a request for evidence; it is a consequence of the dependence relation itself.
The form of the inference
- The B-properties supervene on the A-properties
- Cases 1 and 2 are claimed to differ in some B-property
- Then cases 1 and 2 must differ in some A-property
- No such A-difference can be produced
- Therefore either the claimed B-difference is not real, or the supervenience claim is false
The argument
Proposition The two units cannot be rated differently.
- Rateable value supervenes on the physical and legal facts about a property: two properties alike in every such respect cannot differ in value.
- Units 7 and 9 are claimed to differ in rateable value by eighteen per cent.
- That requires some difference in the base facts — area, aspect, condition, entitlements, access, title.
- The units are mirror-image plans in the same block, same floor, same aspect, same title, same car space entitlement, same condition at the last inspection.
- The valuer, asked for the base difference, cites only that unit 9 sold more recently and for more.
Therefore A recent sale price is evidence about value, not a base fact that value depends on, so no supervenience base for the difference has been produced and the two units cannot properly be rated differently.
The argument never disputes the valuation directly. It asks what the difference could consist in, and the answer offered turns out to be evidence rather than a difference in the underlying facts.
What it is good for
Any domain where an upper-level property is agreed to depend on a lower-level base — value on physical facts, moral properties on natural ones, mental states on physical states. It is the standard argumentative form in philosophy of mind and metaethics.
Where it fails
Supervenience is weaker than it looks. It says the levels covary; it does not say why, and it is compatible with the upper level being fully reducible, wholly independent but perfectly correlated, or anything between. Kim's later work argues it is too weak to state physicalism at all, which was the job it was invented for.
It also depends on getting the base right. Fix the base too narrowly and false differences appear; fix it too broadly — including everything about the whole world — and the claim becomes trivially true and useless.
Common misconceptions
That supervenience means reduction. It does not: two properties can covary perfectly without either being definable in terms of the other, which is exactly why the notion was attractive to people who wanted dependence without identity.
Where it came from
R. M. Hare uses it in ethics in 1952 to argue that two acts alike in all natural respects cannot differ morally. Davidson brings it to philosophy of mind in 1970, and Jaegwon Kim spends thirty years working out what it does and does not deliver.
What became of it
Enormously popular in the 1980s as the way to be a physicalist without being a reductionist, then substantially deflated by Kim's own conclusion that it is too weak for that role.
Lasting influence
It structured the mind-body debate for a generation and remains the standard framing of the relation between the moral and the natural.
Why it still matters
Alive, more modestly than it was.
Put another way
Two identical photographs cannot differ in what they show.
Texts
Jaegwon Kim, Supervenience and Mind
The collected papers in which he builds the notion up and then argues it cannot carry the weight put on it.
Where to start
Hare's original ethical use in The Language of Morals is the clearest and shortest statement of the intuition.
The surprising bit
The philosopher who did most to develop it concluded that it cannot do the job it was introduced to do, and said so at length.
In their own words
There can be no difference of a moral kind without a difference of a non-moral kind
Who held it
G. E. Moore
Connections
- Relies on
- Modus tollens
- Parallel to
- Grounding
The verdict
A precise statement of dependence that turns out not to explain the dependence it states.