Part I: The Problem of Universals

1
Part I The Problem of Universals
2
Types or Categories of Entities
Entities
Universals (Attributes)
Particulars (Things)
Kinds
Relations (Diadic, triadic, polyadic
predicates, to the right of, between,
equidis- tant from etc.)
Properties (monadic predicates like
colour, shape, cha- racter traits )
Biological species (dog, cat, rose, human, or any
specific life form or unified way of being
or basic physical kinds (elements like earth,
water, fire, air)
Genera (animal, plant, ele- ment)
3
The Phenomena to be Explained

• (1) Attribute Agreement
• Many individual items agree in being (a) yellow,
  triangular, courageous, etc. (b) in being
  geraniums, oak trees, human beings, etc. (c) in
  being the parent of, being between, being
  equidistant from, etc., i.e. different things
  agree in possessing (a) the same property or (b)
  belonging to the same kind, and different pairs,
  triples, n-tuples of objects agree in relation.
• (2) Abstract Reference
• Our language abounds in propositions that employ
  abstract terms like courage, wisdom,
  triangularity, circularity that appear to
  refer to, i.e. be proper names for, universals.
  The problem is understanding whether we must
  posit the existence of universals in order to
  understand how these propositions can be true.
• (3) The Truth of Subject-Predicate Sentences
• The truth of sentences like Socrates is wise
  seems to involve some sort of correspondence
  between the sentence and the items that it is
  about. The question is whether the existence of
  the particular entity referred to by the subject
  term is enough, OR whether the existence of the
  multiply exemplifiable entity referred to by the
  predicate term must likewise be posited, so that
  the truth of the proposition consists in the two
  entities being related to each other in the
  manner asserted in the proposition.

4
The Modes of Explanation

• (1) Two-category, realist theories
• A satisfactory explanation of attribute
  agreement, abstract reference, and the truth of
  subject-predicate propositions requires us to
  posit two separate categories of entities,
  universals (multiply exemplifiable, repeatable or
  recurrent entities) as well as particulars
  (entities that occur in one unique place and
  time)
• (2) One-category, nominalist theories
• A satisfactory explanation of attribute
  agreement, abstract reference, and the truth of
  subject-predicate propositions can be given in
  terms of particulars alone, OR no explanation is
  required in that the phenomena in question are
  irreducibly basic phenomena that we cannot get
  under.

5
Three Senses of Metaphysical Realism

• Realism is opposed to (1) idealism, (2)
  nominalism, and, more recently, to (3)
  anti-realism.
• (1) realism is the view that material objects or
  external realities exist (and are what they are)
  apart from (or independently of) our knowledge or
  consciousness of them. Its opposite, idealism, is
  the view that being is dependent of the knowing
  of it (cf. Hirst, Encycl. of Philosophy)
• (2) realism is the view that there are, in
  addition to the familiar concrete particulars of
  everyday experience, universal entities that are
  capable of being instantiated in many
  particulars. Some realists hold that universal
  exist only in the things that instantiate them
  (Aristotle), others that they can also exist in
  their own right or uninstantiated (Plato).
  Nominalism denies of the existence of any but
  particular, individual entities.
• (3) realism is the view that there are truths for
  which there is no evidence and propositions
  backed by evidence which are not true
  anti-realism holds that there is a close tie
  between truth and evidence such that (a) nothing
  is ever true unless there is evidence for it and
  (b) nothing could be backed by a certain kind of
  evidence without being true.

6
Varieties of Realism in sense (2)

• What sort of universals are there?
• (1) Universalia ante res (Platonistic universals
  existing independently of particular things) cf.
  Russell
• There is a different universal for every
  distinct meaning or use of a predicate term or
  abstract universal, whether or not those
  universals are instantiated in space and time.
• (2) Universalia in rebus (Aristotelian
  universals existing only as instantiated in
  particular things)
• (a) There are universals only for basic or
  simple predicates or abstract terms there are
  not additional universals for each composite or
  compound predicate or abstract term to refer to.
• (b) There are universals only for those
  properties, kinds, and relations that our best
  scientific theories posit as real (e.g. muons,
  phuons, quarks. What universals there are is an
  empirical matter for scientists to decide. cf.
  Armstrong

7
Varieties of Nominalism (Particularist
Ontologies)

• (1) Austere Nominalism (cf. Price, Quine) there
  exist only concrete individuals like the familiar
  particular living and non-living material
  objects.
• This is all we need to account for (i) attribute
  agreement (a brute fact that cannot be explained
  by anything more basic), (ii) the truth of
  subject-predicate propositions (explained by
  certain facts about concrete individuals and
  their relations) and (iii) abstract reference
  (only apparently a reference to universals,
  really a special device for talking about many
  concrete particulars at the same time, i.e. about
  all the individuals possessing a certain concrete
  or abstract, monadic or polyadic attribute)
• (2) Trope Theory (cf. Williams) in addition to
  concrete individual things there are also
  concrete individual attributes or properties
  (like colours, shapes, and character traits) that
  resemble each other (trope similarity is a brute
  metaphysical fact that cannot be explained in
  terms of anything more basic).
• This gives a better account of (i) attribute
  agreement (each individual possesses its own
  trope and all the tropes resemble each other),
  (ii) subject-predicate propositions (the
  individual designated by the subject possesses
  one of the resembling tropes correlated with the
  predicate term, and (iii) abstract reference
  (abstract general terms pick out a set or class
  of resembling tropes, e.g. reds, courages,
  etc. )

8
Bertrand Russell, The World of Universals

• 1.Introduction of the problem there are (in
  addition to physical things, mental things, and
  sense data) relations
• 2.Exposition of Platos theory of Forms, and
  replacement of the term Form by Universal
• 3.The logical origins of the problem of
  universals any meaningful sentence must include
  at least one universal, be it a common noun, an
  adjective, a preposition (denoting a relation),
  or a verb.
• 4. The consequences of ignoring relations
  (prepositions) (a) monism of Spinoza and
  monadism of Leibniz, (b) Berkeleys and Humes
  failure to see the obvious refutation of their
  rejection of abstract general ideas.
• 5.Relations do not have mental existence (not
  ideal in the manner of mental acts)
• 6.Relations have being or subsistence, while
  mental things (including sense-data) and physical
  things have existence.

9
H.H. Price Universals and ResemblanceNote
numbers correspond to unnumbered sections key
distinctions are highlighted

• 1. Qualities, relations (including internal
  relations or structure) are universals Platonic
  and Aristotelian theories of universals
• 2.Misleading features of the (Aristotelian)
  Philosophy of Universals (PU) and remedy
  recurrent characteristics instead of
  universals
• 3. First difficulty for PU it cannot account for
  degrees of resemblance (and of instantiation of
  concepts). Objection employs the intensity-extent
  and exactinexact distinctions.
• 4. (a) First classical objection to the competing
  Philosophy of Resemblances (PR) and (b) reply to
  it using the in respect of towards
  distinction
• 5. (a) Second classical objection to the PR (
  Russells refutation of Berkeley and Hume) and
  (b) four replies to it
• 6. Reply to objection against PU in 3. above (see
  determinabledeterminate distinction)
• 7. Conclusion neither PU nor PR has been
  refuted both have their strengths and defects

10
Analysis of Price, 5 (see p. 36)

• O(bjection) to PR Surely resemblance itself is
  a universal, present in many pairs or groups of
  resemblant objects? It is, of course, a relation
  In their attempt to get rid of universals, the
  Philosophers of Resemblance seem to concentrate
  their attention on universals of quality and say
  nothing about universals of relation. (Sounds
  exactly like Russell)
• R(eply) 1 the objection begs the question by
  assuming that in using general words like
  resemblance we commit ourselves to the
  existence of general things (here the universal
  Resemblance), when thats the very thing to be
  proved.
• R2 In declaring resemblance itself to be a
  universal, the objection overlooks the difference
  between (a) the many first-order resemblances
  (qualities and relations) and (b) the
  second-order resemblance among these resemblances
  that explains them all. In other words, having
  begged the question by assuming the things under
  (a) to be universals, it goes on to make the same
  assumption about (b), thus overlooking an
  important difference. (Note first-orderhigher-or
  der distinction)
• R3 In taking resemblance to be a relation (like
  others), the objection overlooks the fact that
  resemblance is (for the PR) more fundamental than
  any relation, not itself a relation, because it
  explains all relations.
• R4 The objection that resemblance is a relation
  like other relations (and that in admitting it we
  admit a universal) leads to contradiction since,
  in the PU itself, resemblance is either not a
  relation at all, or a relation of a very special
  sort

11
Quine, On What There Is

• 1. Platos beard (or the riddle of non-being
  what is not must be since otherwise it would be
  meaningless to say that it is not) and the
  difficulties into which it leads anyone who
  wields Occams razor (takes the negative side in
  an ontological dispute) what is alleged not to
  exist (in space and time) must still exist
  somehow, be it (1) as an actual idea existing
  only in the mind (McX) or (2) as an unactualized
  possible subsisting eternally outside it (in
  Platos heaven?), for which it is hard to give
  identity conditions and if (2), the further
  consequence that unactualized impossibles
  (contradictions) likewise exist can only be
  avoided by asserting them to be meaningless, and
  this makes a general criterion of meaninglessness
  impossible, since there can be no generally
  applicable test of contradictoriness (Wyman).
  (pp. 425)
• 2. Russells theory of definite descriptions as
  a means of shaving off Platos beard by
  analyzing out particular names like Pegasus
  (for which definite descriptions may be
  substituted) in the contexts (i.e. sentences) in
  which they occur, so that we can say (e.g.)
  Pegasus is not without even appearing assert
  (self-contradictorily and meaninglessly) that the
  thing which is not is. (pp. 457)

12
Quine, On What There Is (cont.)

• 3. The gulf between meaning and naming (or
  between sense and reference). This rules out the
  inference the name x is meaningful (in such
  and such a use) therefore it names something, x
  therefore some x exists. (pp. 478)
• 4. The ontological problem of universals the
  ontology of universals admits occult entities
  like attributes, relations, classes, numbers,
  functions and meanings the ontology of
  particulars holds that words and propositions are
  significant or meaningful, and that this is an
  ultimate and irreducible fact about language that
  does not require or admit of explanation in terms
  of universals like meanings. (pp. 4850)
• 5. Ontological immunity and ontological
  commitment To be is to be the value of a bound
  variable. (pp. 501)
• 6. Parallels between the realist, nominalist, and
  conceptualist positions the medieval controversy
  about universals and logicist, intuitionist, and
  formalist positions in contemporary philosophy of
  mathematics (pp. 522)
• 7. Reasons for shifting from the ontological to
  the semantic plane and reflexion on the
  relativity of ontology (pp. 52end)

13
Quines Two Chief Lessons

• 1. Regarding singular names and the problem of
  Platos beard What seem to be names (e.g.
  Pegasus) which, to be meaningful, must refer to
  existing individuals are, correctly understood,
  incomplete symbols (Russell) the sentences in
  which they occur can be translated into others
  containing only non-referring expressions, i.e.
  (1) individual variables, (2) predicate
  variables, and (3) bound variables (of
  quantification like something, nothing,
  everything).
• 2. Regarding the ontological problem of
  universals or abstract entities (like
  attributes, meanings, species or classes,
  numbers, and functions) First, attribute
  agreement may taken as an ultimate and
  irreducible matter of fact that needs no
  explanation by occult entities like universals.
  The same applies to meanings we can avoid
  committing ourselves to the existence of meanings
  as something that that utterances have (or
  have in common) by saying instead that the
  utterances are significant (meaningful) or
  synonymous (and, ideally, by explaining these
  words in terms of behaviour). Next, as regards
  species, we are ontologically committed to the
  existence of anything over which our bound
  variables range (until such time as we can show
  that our bound variables, like names, are just an
  avoidable manner of speaking). Finally, regarding
  numbers and other mathematical entities, here
  too an extravagantly platonizing philosophy of
  mathematics (formalism), and even a moderately
  platonizing one (intuitionism) are open to
  challenge from one (formalism) that regards
  mathematical symbols as no more than useful
  conventions involving no ontological commitment.

14
Williams, The Elements of Being

• 1. The cardinal problem of ontology substance
  (traditionally individual, different, many,
  particulars) and attribute (traditionally
  universal, same, one) or better
• Substances and Universals (note substances or
  particulars are really the next topic right now
  it is really the question of whether or not
  universals exist or not)
• 2. The alphabet of being (the elements of
  being)
• First Proposal regard inexactly similar
  properties (colours, shapes, sizes, tastes, etc.)
  not as abstract general kinds (universals), but
  as themselves constituted by exactly similar, yet
  numerically distinct abstract individual parts
  and regard these abstract particular parts or
  tropes as the ultimate entities, (and therefore
  the only true entities) making up both concrete
  individual substances (and their concrete parts)
  and classes or universals.

15
Williams, The Elements of Being (cont.)

• 3.The syllabary of being (ways in which the
  elements are combined)
• Second Proposal there are two and only two ways
  in which tropes are combined
• 3.1. Location (at the same place at the same
  time), to form individuals or particular
  substances (concurrence sums of tropes), i.e.
  individuals are reducible to these tropes.
• 3.2. Similarity, to form universals or
  classes as sums of similar tropes or
  similarity sets of tropes, i.e. universals can
  be eliminated, leaving only similarity sets of
  tropes.
• 4. Dispelling the mysteries of predication
  (subject-predicate propositions) and
  existence/essence (existing things and their
  natures).

16
Re Tropes as Abstract Particulars

• Question
• What are a babys first experiences of?
• Possible answers
• 1. concrete particular things like the ball in
  its crib
• 2. abstract universals or essences like redness
  in general and roundness in general (which it
  will later come to conceive as a ball)
• 3. abstract particulars or abstract parts
  (tropes) of the ball like this particular red (of
  this ball here), this particular round(ness of
  this ball here) etc.
• An example of a trope, then, is a particular red
  sense datum, which admittedly has such
  distinguishable components as a shape and a size
  as well as colour and involves the attributes
  of hue, brightness, and saturation still it is
  an abstract part in comparison with a whole
  coloured solid e.g. a ball (from latter part of
  Williams paper, not included in our excerpt). In
  general, we can say that tropes are abstract
  parts like colour, shape, position, actions
  performed, locations of a thing as opposed to its
  concrete parts (e.g. top and bottom, right and
  left sides, constituent body parts, and so on.
• Other names for tropes first accidents (in
  medieval terminology), unit properties, cases,
  aspects, property tokens, particularized
  properties, perfect particulars, abstract
  particulars.

17
Williams, The Elements of Being (cont.)

• Questions for Discussion
• Q1 Name the circumstance that gives rise to
  theories of subsistence and inherence (the fact
  about the world to be explained metaphysically).
• Q2What might a realist metaphysical explanation
  of this circumstance be like?
• Q3 W. is here staking out a nominalist or a
  realist position? How does his nominalism differ
  from Prices?
• Q4 Why does W. give his lollipops and their
  parts funny names? Whats he trying to get across?

18
Williams, The Elements of Being (cont.)

• Q5Whats W.s point about the word same in
  (1) Ryan and his brother have the same nose
  and (2) Ryan and his brother have the same
  father.
• Q6 Parse the sentence about Napoleon on p. 60.
  Why are the tropes tropes, and the non-tropes
  non-tropes?
• Q7How does W.s attempt to reduce (actually
  eliminate) particular things and universals
  (shared properties and relations) from the basic
  furniture of the universe work?
• Q8 We wouldnt speak of a universal as being a
  particular (individual) thing (although Platos
  Forms seem to be described this way). What of
  particulars that are abstract rather than
  concrete?

19
Armstrong, Universals as Attributes

• I. The Principle of Instantiation (PI). Arguments
  for universalia in rebus (properties) and inter
  res (relations), but against uninstantiated,
  transcendent, other-worldly universalia ante res
  (Platonism)
• (a) if not PI, then things have a blob-like
  rather than layer-cake structure (?)
• (b) onus of proof argument onus on Platonists
• (c) fallaciousness of traditional Argument from
  Meaning (or One-Over-Many Argument)
• (d) fallaciousness of traditional Falling-Short
  Argument (or Argument from Relations)
• (e) Tooleys subtler argument (?) inconclusive
• II. Two Arguments against Disjunctive and
  Negative Universals
• (a) Universals should have something in common
  disjunctive and negative universals (like
  C-or-M and not-C) may not.
• (b) Universal properties are associated with
  powers a further disjunct or negation both blurs
  identity conditions and adds nothing to power
• (c) Note The same argument does not hold
  against conjunctive universals

20
Armstrong, Universals as Attributes

• III. Family Resemblance. The truth that sometimes
  a family resemblance rather than a shared
  property underlies a universal predicate does not
  imply There are no universals (nominalism),
  but only There is not a universal for every
  predicate (realism). Family resemblance like
  shared properties and common relations must be
  explained in terms of universals. Which
  universals? Armstrong The manifest or
  qualitative image of the world is best accounted
  for in terms of the quantitative monadic and
  polyadic universals (properties and relations)
  posited by physics.
• IV. The Truth-maker Principle. More is involved
  in the truth of contingent subject-predicate
  propositions than just (a) the existence of the
  particular subject and (b) that of a predicate
  universal, namely (c) the existence of states of
  affairs in the world as the ontological ground
  (truth-maker) of contingent truths. States of
  affairs are required by Class Nominalists and
  Resemblance Nominalists (though not by Predicate
  Nominalists). States of affairs (if they exist)
  have a non-mereological composition since
  different states of affairs can have the same
  components (unlike the part-whole relation).

21
Armstrong, Universals as Attributes

• V. Properties and universals occur only within
  states of affairs as their (multiply-instantiated)
  constituents No Bare Particulars (understood
  as a particular occurring outside states of
  affairs, a particular that does not instantiate
  any properties or relations).
• VI. Solution to the Antinomy of Bare Particulars
  (Quilter). Distinguish Thin and Thick,
  Particulars. Thin particular that something
  (substratum) in an entity which is linked to its
  properties by instantiation (yet not either
  identical with them or capable of existing
  unlinked). Thick particular thin particular
  plus all its properties state of affairs.
  Intermediate particular thin particular and
  some of its properties.
• VII. Univerals as Ways to guard against
  conceiving univerals as existing uninstantiated
  (substantializing them) or particulars existing
  without properties/relations think of universals
  as ways in which particulars are.

22
Armstrong, Universals as Attributes

• VIII. A world consisting of states of affairs
  constituted by properties and universals as the
  solution to (1) the (Platonic) Multiple Location
  Problem
• IX. And (2) the problem of higher-order types or
  universals, like properties and relations of
  properties and relations, which are no longer
  needed, except to account for the Laws of Nature
  (relations between two properties) and functional
  laws.
• X. Formal properties of the resemblance relation
  (symmetry, transitivity of exact resemblance,
  non-transitivity of inexact resemblance), which
  are explanatory ultimates on the resemblance
  theory of universals, are explained by the
  universals theory
• XI. The resemblance relation between universals
  can be eliminated.
• XII. Instantiation as a primitive, inexplicable
  explanatory concept
• XII. The apparatus of the universals theory
  (i) properties and relations, (ii) states of
  affairs instantiation as a fundamental tie.

23
Varieties of Realism and Nominalism Summary
Realism
Nominalism
Universalia post res after particulars i.e.
added by the mind, which (a) forms universal
concepts or picks out classes (predicate or
conceptual nominalism), or (b) picks out
resemblances (resemblance nominalism)
Universalia in rebus (in particulars) No
uninstantiated universals Aristotelian One-World
Doctrine structured by immanent universals or
forms
Universalia ante res (before particulars)
including uninstantiated universals or
Forms Two-world Doctrine of Transcendent
Universals or Forms (Platonism)
24
Part Two The Structure of Particulars
25
The Ontological Structure of Particulars
Alternative Theories
A. Reductivist Theories
B. Anti-reductivist Theories

3. Substance Theories (Things are basic entities
capable of ontological analysis but not in terms
of more basic constituents)
1. Bundle Theories (Things are reducible to
bundles of properties)
2. Substratum or Bare Particular
Theories (Things are reducible to bundles of
properties plus a bare substratum)
1.( a) Realist individuals or substances are
bundles of universals or identical properties
co-instantiated with other properties in
different bundles cf. later Russell (BTU)
1. (b) Nominalist or trope-theoretical Individua
l things are bundles of tropes, i.e. uniquely
individual but similar properties cf. Berkeley,
Hume, Ayer, and Williams (BTT)
2. (a) Realist an individual entity includes
something over and above the universal
proper-ties it exemplifies, namely a propertiless
(bare) substratum or bearer cf. Locke, early
Russell, Bergmann, Allaire, Black (BPU)
2. (b) Nominalist or trope-theorist an
individual entity is something over and above
its tropes, namely etc. Cf. Arm-strong (BPT)
3. Aristotelian or quasi-Aristotelian What
possesses the attributes is not a bare
substratum, but the familiar particular or
individual (organic) thing. Natural kind is a
basic attribute, but not a constituent cf.
Aristotelians, Van Cleve (AST)
26
The Ontological Structure of Particulars
Alternative Theories Some Abbreviations

• BTU The reductionist bundle theory according to
  which properties are universals instantiated in
  multiple different particulars (one-category
  theory)
• BTT The reductionist bundle theory on which
  properties are tropes unique to each individual
  (one-category theory)
• BPT The reductionist theory that posits a
  substratum (a bare or property-less
  particular) as the bearer of properties, whether
  the latter are universals or tropes
  (two-category theory)
• AST anti-reductionist Aristotelian substance
  theories

27
The Ontological Structure of Particulars
Alternative Theories Some Abbreviations

• PCI The principle of constituent identity
  identity of constituents entails numerical
  identity of the constituted whole (axiomatic for
  BTU, BTT, and BPT, but irrelevant for AST)
• PND The principle of the discernibility of
  non-identicals (all non-identicals have some
  unshared property)
• PII The principle of the identity of
  indiscernibles (all indiscernibles things
  having all the same attributes are identical)
• BBT The bundle-bundle theory according to
  which individual things are bundles (series in a
  definite relation) of momentary things which are
  in turn bundles of properties.

28
Reductivist and Anti-reductivist Theories

• Reductivist theories
• take individual entities to be wholes
  constituted by more basic entities, such that the
  constituent is ontologically prior to (more basic
  than) the whole. Thus, it is a shared assumption
  of the BTU, the BTT, and the BPT that PCI
  (Principle of Constituent Identity numerical
  identity of constituents entails numerical
  identity of the constituted whole) or (by
  contraposition) numerical difference between
  constituted entities entails some difference in
  their constituents (though not necessarily in
  their attributes, unless they have no other
  constituents, something substratum theorists
  deny).
• Anti-reductivist theories
• take complex individual entity of certain kinds
  to be absolutely, i.e. irreducibly basic such
  that its (universal) accidental properties or
  attributes (what it is like) can only be
  understood on the basis of its necessary
  attributes (what kind or species of entity it is
  and the properties that follow therefrom). The
  wholeconstituent model and the PCI (both
  borrowed from science and inappropriate to
  ontological analysis) are rejected in favour of a
  natural kindaccident model of structure.

29
PDN and PII

• The Principle of the Discernibility of
  Non-identicals
• Numerically different things must differ in
  nature, that is, in respect of at least one
  attribute, must have some unshared attribute.
• Thus All non-identicals (any two things) must
  be discernible (or if different in number, then
  somehow different in nature)
• The Principle of the Identity of Indiscernibles
• If what are putatively two things (call them a
  and b) do not differ in any attribute, they are
  really one and the same thing designated by two
  names, i.e. they differ in name only, but not in
  nature and therefore not in number
• Thus (by the contraposition of PDN) All
  indiscernibles must be identical (i.e., if not
  different in nature, then not really different in
  number, but only in name).
• Or in logical terms Necessarily, for any x, y,
  and P, where x and y are any concrete objects and
  P is any property, if P is a property of x if and
  only if P is a property of y, then x and y are
  identical. Or, the same thing in symbols
• ?(x, y, P)(Px Py) ? x y

30
Why does the PII matter?

• The BPT (substratum theorist) argues that since
  (i) the BTU entails the PII, and since (ii) the
  PII is false, (iii) BTU must be false.
• Re (i)
• The BTU combines two theses
• (a) Properties or attributes are universals and
  hence identically the same in all the individuals
  that jointly instantiate or exemplify them
  (realist thesis properties are universals)
• (b) Properties or attributes are the only
  constituents of individuals, i.e. individuals are
  nothing but bundles of universal properties.
• Now (a) and (b) together with the Principle of
  Constitutent Identity (PCI same
  constituents/same entity principle) entail that
  two things having all he same attributes are one
  and the same thing, or (better) there cannot be
  two things having all the same attributes (PII).
• If PII is false, then either (a), (b), or the PCI
  must be false. But the realist thesis and the PCI
  are both clearly true, therefore (b) (and hence
  the BTU) must be false.

31
Different Versions of the PII

• Weakest (empirical) version (Some unshared
  attribute in fact)
• As a matter of contingent fact, no two things
  have all their properties or attributes in common
  (Russell, Casullo)
• Standard version (Some unshared attribute
  necessarily)
• Necessarily, no two things can have all their
  properties in common, i.e. it is metaphysically
  impossible for there to be diverse yet
  indiscernible objects (Leibniz)
• Very Strong version (No shared attribute
  necessarily)
• Necessarily, different things can have no
  property in common, i.e. they must differ with
  respect to all their properties (Spinoza, for
  reasons having to do with his conception of a
  thing or substance, A.J. Ayer and Hume, the
  latter two being defenders of the BTT)

32
Black, The Identity of Indiscernibles

• 1. As First Argument for the PII using being
  identical with (or being different from as
  a property.
• Bs rebuttal trivial. Property must be
  interpreted more narrowly to avoid triviality.
• 2. As Second Argument for the PII using the
  verificationist theory of meaning if the PII is
  not true, any number of items could be called
  different (e.g. B has ten different hands)
  since there need be no discernible difference by
  which to confirm or disconfirm that these items
  are in fact two, three, etc. But if assertions of
  difference are in principle unverifiable, then
  they are meaningless (according to a widely
  accepted criterion of meaning). Hence to assert
  meaningfully that A and B are two, it must be
  possible to discern a property that A has and B
  lacks (or vice-versa) PII

33
Black, The Identity of Indiscernibles

• 3. Bs counter-example demonstration of the
  possibility of (two or more) numerically distinct
  yet indiscernible objects (indiscernible
  counterparts), or of the falsity of PII and As
  constant retort that this is unverifiable and
  hence meaningless.
• (Note Since the PII can be formulated (see
  above) as
• ?(x, y, P)(Px Py) ? x y,
• this allegedly necessary truth can be refuted by
  showing that the following is a possible state of
  affairs
• (x)(y) x ? y (P)(Px Py)
• (To be read as There exist two objects, x and
  y, such that x is not identical with y, and for
  any property P, P is a property of x if and only
  if it is a property of y.)

34
Allaire, Bare Particulars

• 1. Aim (114) to Reconcile the Individual-Characte
  r Analysis of Subject-Predicate Propositions
  (BPU) and the Principle of Acquaintance, a pet
  principle of empiricism which states (roughly)
  anything posited as a basic entity or referent of
  an indefinable term must be immediately given in
  experience.
• 2. The Dialectics of Sameness and
  Difference(115-17) both the nominalist BTT and
  realist BTU analyses of two similar but
  non-identical red discs face insuperable
  objections that do not affect the
  Individual-Character Analysis (BPT). For unlike
  the BTU and BTT theories, the BPT accounts for
  the facts that (a) there is something that is the
  same and (2) there is something that is
  different.
• 3. Dialectics versus Phenomenology (117-8)
• 4. Solution (118 middle-120) Allaires attempt
  to show phenomenologically that, unlike the case
  of Bergmanns red spot, where it is hard to see
  that a bare particular is immediately given,
  there is no problem about our just seeing this
  (in the manner of direct acquaintance) in the
  case of two exactly similar red discs.

35
Van Cleve, Three Version of the Bundle Theory

• I. Six Objections to (1) the crude version if
  a thing is nothing but a set of properties
  (BTU), then (O1) there exists an actual thing for
  every actual set of properties (O2) actual
  things exist necessarily since properties do
  (i.e. must be instantiated) (O3) even sets of
  two members must be said to exemplify their
  members, i.e. the doubleton redness,
  roundness must be red (O4) things cannot change
  properties since sets cannot change members (O5)
  all properties of a thing are essential to it
  (O6) the PII is a necessary truth. But the
  consequences in (O4)(O6) are all false. Hence
  BTU false.

36
Van Cleve, Three Version of the Bundle Theory

• II. The (2) sophisticated version of BTU (a
  thing is a set of properties that are
  co-instantiated) evades (O1)(O3). Thus, (O1) a
  set of properties constitute a thing only if
  co-instantiated (O2) any thing is capable of not
  existing if its properties, which necessarily
  exist, cease to be co-instantiated and (O3)
  doubletons need not exemplify their members as
  things do, since things do owing to the special
  relation of co-instantiation which is more than
  membership in a set. But it cannot evade O4O6
  (despite the efforts of Loux and others to show
  that it can).

37
Van Cleve, Three Version of the Bundle Theory

• III. The new bundle theory that avoids the
  objections (O4)(O6) is modelled on the new
  phenomenalism it does not populate the world
  with individual that are O4 incapable of
  change, O5 devoid of accidental properties, and
  O6 qualitatively unique because it does not
  populate the world with individuals at all, but
  eliminates individuals in favour of properties
  that are co-instantiated at one or more places in
  space.
• IV. Although Ayer has come close, no one has ever
  held this theory which, by eliminating physical
  things altogether, would also eliminate any human
  being that holds it. Thus, all three versions
  fail. The alternative is a theory involving
  substance (though not as a bare particular).

38
Questions for Discussion

• Q1 Speaking of the BTUs appeal to impure
  properties like being self-identical in defence
  of the PII, VC says (on p. 122) One cannot have
  it both ways. Is the defence in question
  circular, or is it self-contradictory?
• What about the other circularity issue (p. 123)?
• Q2 How do Loux and Russell try to secure the
  bundle theory against the objection that all
  subject-predicate propositions are tautologies
  (and all predicates necessary) since any
  predicate (property) is always already contained
  in the subject (bundle)?
• Q3 Describe the core-bundle approach to
  restoring a distinction between necessary and
  accidental properties. Is it more promising? What
  about the WIP-property approach?
• Q5 Thoughts on VCs account of Leibnizs reasons
  for holding the PII?

39
Casullo, A Fourth Version of the Bundle Theory

• Introduction Since there is a fourth
  (Russell/Bergmann/Casta-neda) version of the BT
  that is immune to the last three objections
  (renumbered O1,O2,O3) Van Cleve (VC) brings
  against the second version, the BT is still
  unrefuted.
• I. Step (i) Summary of VCs last three
  criticisms of the second version on which (1) A
  thing is a bundle of co-instantiated properties.
• Step (ii) Distinction of the problems of (a)
  individuation and (b) identity across time. (a)
  (b) are solved together by the AST, but
  separately by both the BPT and the fourth
  version of the BT BBT (the two-tiered
  bundle-bundle theory)
• Step (iii) Replacement of the BT that holds (1)
  with a BBT which affirms the joint theses (1)
  A momentary thing is a bundle of co-instantiated
  properties and (2) An enduring thing is a series
  of momentary things in some contingent relation
  R.
• Step (iv) Demonstration that BBT, unlike BTU, is
  not susceptible to O1 or O2.

40
Casullo, A Fourth Version of the Bundle Theory

• II. Demonstration that BBT, unlike BTU, is not
  susceptible to O3 (BTU?PII counter-examples show
  -PII therefore BTU)
• Step (i) Distinguish strong (SBT) and weak
  versions (WBT) the latter is not committed to
  the necessary truth of PII, but only to (3) No
  two things in fact have all properties in common.
• Step (ii) Substitute for O3 a weaker version of
  the objection called O4 to the effect that (3) is
  empirically unsupported.
• Step (iii) what makes 04 seem a plausible
  objection to (3) is that the reductivist WBT
  cannot appeal to their unique spatial and
  temporal properties in order to individuate
  particulars without circularly (or
  inconsistently) appealing to irreducible things
  as spatio-temporal reference points. In other
  words, spatial and temporal properties are not
  pure properties, and (4) No two things in fact
  have all pure properties in common is empirically
  unsupported.
• Step (iv) Way out The WBT can be further
  weakened by substituting (5) a small number of
  things for (4) no two things and (5) is
  empirically well supported.

41
Casullo, A Fourth Version of the Bundle Theory

• III. Consideration of O3 as the only objection
  capable of refuting the SBT now that O1 and O2
  have been shown to have no force against it.
• Step (i) The standard form of O3 is refutation of
  PII by counter-example.
• Step (ii) Consideration of whether
  counter-examples of PII are possible if (as
  Russell and Goodman argue) position in physical
  space is among the pure (monadic) properties of a
  thing, then in order to imagine two objects one
  must imagine them as occupying different places
  in the visual field, and counter-examples like
  Blacks fail.
• Conclusion Even the SBT remains unrefuted.

42
Part Three Causality
43
Competing Theories
Types of Theory of Causality
2. Modal Theories Committed to Causal Necessity
or Necessary Connection
1. Non-modal or Eliminative Theories (No
Necessity or Necessary Connection involved in
the Causal Relation)
(a) Regularity Theories (Hume, J.S. Mill, Mackie)
(e) Intuitionist Theories (Kant)
(b) Counter-factual Analysis (Lewis)
(d) Entailment Theories (Ewing)
(c) Direct Observation Theory (Anscombe, Armstrong
)
44
Types of Necessity
necessity

• de re necessity
• necessity of things
• real necessity (in the world)

• de dicto necessity
• necessity of propositions
• of necessary truths (and
• falsehoods)
• logical necessity
• (or impossibility)

de re metaphysical necessity
de re causal or natural necessity
psycho- logical mind-mind causal interactions
connections
physical body-body causal inter- actions
and connections
psycho-physical mind-body causal interactions
and connections
necessary properties (essentialism)
a necessary being (God) (theism)
45
Ewing, Cause

• 1. The Empiricist Regularity Theory and Three
  Criticisms
• The Humean theory (that cause and effect are
  nothing but a particular instance of a regular
  sequence of events similar to the cause followed
  by events similar to the effect) cannot account
  for
• (a) the causes of unique events
• (b) the distinction of (i) those regular
  sequences that are merely casual (accidental
  correlations) from (ii) those that are genuinely
  causal
• (c) rationality, freedom, and memory (for
  beliefs to be rational they must be determined to
  exist by, and not just follow upon, a
  consideration of reasons for actions to be free,
  they must be determined to occur by the will and
  for memories to be reliable they must be
  determined to exist by the events remembered and
  by nothing else).

46
Ewing, Cause

• 2. The Rationalist or Entailment Theory
• The theory that the effect follows necessarily
  from the cause in a manner analogous to that in
  which the conclusion of an inference follows from
  the premise has four arguments in its favour and
  three against.
• In its favour
• (a) we can (i.e. are in fact justified) in
  making causal (inductive) inferences this makes
  sense if the conclusions are somehow entailed by
  the evidence
• (b) that things behave in certain ways does not
  seem to be a coincidence (on a cosmic scale), but
  to have reasons and the reason seems to be that
  such behaviour is somehow entailed by their
  natures.
• (c) we have direct, positive insight into a
  quasi-logical connection in psychological matters
• (d) we have no other way to justify the
  rationality of induction

47
Ewing, Cause

• Arguments against the Entailment View (with
  Ewings rebuttals)
• (i) we do not have any positive insight into a
  logical or quasi-logical connection between
  events (rebuttal we dont have a positive
  insight that there isnt one either it may be
  there, but as yet undiscovered, as are many
  actually existing logical connections in
  mathematics, for example)
• (ii) effects succeed their causes in time, while
  logical relations of entailment between
  propositions are timeless (rebuttal doesnt
  suffice to disprove a necessary connection of a
  temporal sort, if there is independent evidence
  for it)
• (iii) causal reasoning gives probability,
  logical reasoning certainty (rebuttal lacking
  direct insight into the full cause, we make
  inferences that are only probable. But the full
  cause may still entail the effect with perfect
  necessity/certainty).

48
Questions for Discussion

• 1. What is the common sense view of causality? Is
  the entailment theory more closely akin to it
  than the regularity theory?
• 2. In what way does the regularity view make
  causation something that can be empirically
  observed? Isnt Humes whole point that nothing
  like a necessary connection can be observed?
• 3. How does the regularity view eliminate the a
  priori.
• 4. What do you think of Ewings direct insight
  argument on p. 279? Is it refuted by the fact
  that we are sometimes wrong about causal
  connections?
• 5. How does the argument from the rationality of
  induction (p. 280) go?

49
Anscombe, Causality and Determination

• 1. Breaking the Tie between Causality and
  Determination
• Step I. The historical connection (Aristotle to
  Russell) between causation and (a) necessitation
  (determination), (b) universality, (c) the DI/DC
  (difference of issue/difference of circumstances
  principle).
• Note the similarity between DI/DC and Humes
  principle SC/SE (Same Cause/Same Effect)
• Step 2. The core idea of causation is
  derivation (this coming from, arising out of,
  having as source that) necessitation and
  universality are super-added ideas what makes
  one think otherwise is Humes thesis that we can
  never observe causality in individual cases, and
  that even in such cases we are really thinking of
  what always happens in cases of the same kind.

50
Anscombe, Causality and Determination

• Step 3. Two replies to Humes thesis (1)
  turning the tables on Hume, it can be replied
  that in Humes sense of observe (the way we
  observe redness, for example) we dont even
  observe physical events (but only our mental
  sense-data) but (2) in the sense in which we
  actually observe physical things and events (e.g.
  fire) we do indeed observe causality.
• Re (b), in assigning a cause we do not think of
  laws in the sense of exceptionless regularity
  under controlled and constant conditions so much
  as the inconstant and uncontrolled conditions
  (different circumstances) under which that which
  the law prescribes does not happen (different
  issue).
• 2. The Innovations of Indeterministic Physics as
  Breaking the Tie

51
Mackie, Causes and Conditions

• 1. The received formula A causes B A is a 1
  necessary and 2 sufficient condition of B and a
  first modification.
• Step 1 Introduction of the notion of an INUS
  condition
• A is at least an INUS condition B A is an
  indispensable (non-redundant) part of some
  complex set of both positive and negative
  ()conditions (e.g. ABC, or, substituting X for
  BC, AX) which are jointly minimally sufficient
  (but not necessary) for B. (Here indispensable
  means that, without A, X would not have produced
  B minimally sufficient means that no
  dispensable or redundant factors are included in
  X and not necessary means that was at least
  one other set of minimally conditions (e.g. DEF,
  or simply Y).
• Thus A is an Insufficient but Necessary part of
  a condition which is itself Unnecessary but
  Sufficient for the result (an INUS condition).

52
Mackie, Causes and Conditions

• Step 2 Full Analysis of A caused B as
• (i) A is at least an INUS condition of B
• (ii) A was present
• (iii) the other factors in X were also present
  (if positive) or absent (if negative)
• (iv) no other set of sufficient conditions, Y,
  was present.
• 2. Further Modification of (i) based on the
  notion of a causal field
• (A is at least an INUS condition B in the field
  F) which solves the problem of the impossibility
  of giving sufficient conditions short of the
  whole prior state of the universe)
• 3. Analysis of General Causal Statements

53
Questions

• 1. Name two sorts of circumstances under which A
  would be more than an INUS-condition, namely a
  necessary condition of B?
• 2. Explain what Mackie means by a condition of
  these forms (AX or Y), (A or Y), AX, A.
• 3. How does Mackie analyze credit restriction
  causes unemployment and eating sweets causes
  tooth decay.

54
Part IV The Reality of Time
55
Theories of Time
Realist Theories
Anti-realist (or Idealist) Theories (Parmenides, P
lato, Spinoza, Kant(?), Hegel Bradley,
and McTaggart)
B-Theorists Time is an eternal frame-work
of before/after relations that serves, along
with space, to situate things
events four-dimensionally and tenselessly, with
no spe- cial status for the present
A-Theorists Time is a non- relational
but tensed property of things events in time,
i.e. that pass from future, to present, to past
Presentists (Descartes?, Prior)
Four-dimensionalist Theorists (Hera- cleitus ?,
Broad, Taylor)
Old Tense- less Theory (Leibniz, Russell, Carnap,
early Smart, Williams Quine)
New Tense- less Theory (later Smart, Mellor)
56
The A- and B- Theories
The A-Theory 1. Transientist or Flux
Theory whereby there is something ontologically
privileged about the present in relation to past
and future 2. Presentism or Four-Dimentionalist
reality consists of the present alone or the
present, past, and future, with a Special status
for the present 3. Anti-reductionist Past,
present, future and tensed verb forms express
objective features of things, events, and time
so that tensed language is irreducible (must be
taken at face value)
The B-Theory 1.
Eternalist/Relationalist/Equirealist Time is an
eternal framework of relations of earlier than,
later than and simultaneous within which every
event and thing has an unchanging (and equally
real) position 2. Four-Dimensionalist Things and
events are spread out in time as well as in
space (in accordance with the best scientific
theories) the regions of time equally
real. 3. Reductivist the predicates
present, past and future and all tensed
descrip- tions are subjective, i.e. relative to
the present of the speaker they can be
translated into the tenseless terms of the
B-theory either through token- reflexive
analysis or the use of dates
57
McTaggart, Time Overview

• Preliminaries Distinction of two ways of being
  in time (1) being in the B- series (a system
  of timeless relations of earlier and later,
  among events) (2) being in the A-series (the
  allegedly necessary passage of each event
  and each moment of time from future to present
  to past)
• I. Stage One of the Argument
• P1 C ? T (if no real change, then time is
  unreal)
• P2 As ? C (if no real A-series, then no real
  change)
• C1 \ As ? T (if no A-series, then time is
  unreal)
• Corollary As ? Bs (if no A-series, then no
  B-series, since the B-series is a set of
  temporal relations and without the A-series
  there is neither time nor temporal relations)
• Consideration of Three Objections each of which
  tries to show that the A-series is inessential to
  time, i.e. that the B-series is sufficient.
• II. Stage Two of the Argument
• P3 ?As (it is logically impossible for an
  A-series to exist)
• C2 \ T (hence time is unreal, i.e. nothing
  that exists can be temporal i.e. really in
  time, although we have no experience that does
  not appear to be temporal)

58
McTaggart, Time P1 is Axiomatic

• It would, I suppose, be universally admitted
  i.e. it is a self-evident axiom that time
  involves change i.e. there could be no time if
  nothing changed and no change if there were no
  time.
• So T ? C (time if and only if change)
• \C ? T
• i.e. if no change, then time is unreal

59
McTaggart, Time P2

• P2 As?C (if no real A-series, then no real
  change)
• This itself is a conclusion of a modus tollens
  argument that runs
• First Premise ?(Bs As Ch) ? ?(Ch As)
• If then the B-series without the A-series can
  constitute time, change must be possible without
  the A-series, i.e.
• Second Premise ?(Ch As)
• But this is impossible since relations of
  earlier and later among events or among moments
  of absolute time are permanent and no event can
  cease to be or alter its position simply as an
  event in the B-series, i.e. change must be change
  from being future to being present or being
  present to being past.
• Conclusion ?(Bs As Ch) i.e. the
  B-series without the A series is not sufficient
  for change, or P2 As?C

60
McTaggart, Time Three Objections

• 1.Objection stemming from Russels B-Theory of
  Time A-series does not exist per se tensed
  statements about past, present, and future events
  (e.g. The battle of Waterloo is in the past)
  are reducible to timeless (or untensed)
  propositions stating that the event is earlier,
  later, or simultaneous with a particular
  utterance in the B-series (e.g. The battle of
  Waterloo is earlier than this judgment).
• Reply Russell rejects the A-series, but believes
  in change. But it has been argued that if no
  A-series, then no change. New support Everything
  about a fact, including its place in the
  B-series, is unchanging. (Example death of Queen
  Anne.) For facts to change, then, they must
  change their place in the A-series.

61
McTaggart, Time Three Objections

• 2. Objection that there are non-existent
  (fictional) series where the events are in
  temporal relations of earlier and later (a
  B-series) but do not change from future, to
  present, to past (not in the A-series) because
  they are do not exist at any time.
• Reply The series is not temporal unless in the
  A-series, and since only things that exist are in
  the A-series, only things that exist are in time.
  Things may be believed or imagined to exist in
  time only if believed or imagined to exist in the
  A-series.
• 3. Objection that there could be several
  different time series instead of just one, so
  that there would not be one present in relation
  to which events were future, then present, then
  past.
• Reply there would still be some present for each
  independent time series for it to be a time
  series at all.

62
Broad, Ostensible Temporality

• McTaggarts philosophical howlers in P3
• 1) treating absolute becoming the A-series as
  if it were a species of qualitative change such
  that the present, past, and future of the
  A-series are incompatible qualities or
  determinates of the same determinable (time)
  which can no more be consistently attributed to
  the same subject that red and blue as
  determinates of the same determinable (colour)
  can be attributed to the same entity (either
  simultaneously or timelessly, though they can be
  predicated of the same thing at different times).
• 2) replacing temporal copulas i.e. is (now),
  was (in the past), and will be (in future) by
  non-temporal copulas is timelessly or eternally
  or sempiternally plus temporal adjectives or
  temporal predicates (presentness, pastness,
  futurity).
• Thus, it will rain there is (timelessly) a
  rainy event and it has the property of futurity.
  And generally x (an event or time) is, x
  was, and x will be x has (timelessly) the
  incompatible temporal properties or qualities of
  presence, pastness, and futurity.

63
Taylor, Time and Eternity

• Introduction The eternalist (Platonic) argument
  against becoming and the datum or appalling
  fact which it denies.
• I. Description of the datum The pure becoming
  (becoming older) of (a) things, (b) events, and
  (c) time itself, whether absolute in the case of
  the now existent or relational in that of the
  not-yet or no-longer existent, as the
  presupposition of all other change in events and
  things.
• II. Rational reflexion on the datum and the
  absurdities it brings to light (1) both
  observable and a priori (2) events are changes
  (ordinary changes in the properties of something)
  and yet themselves change (pure becoming) (3)
  this pure becoming seems to requires the
  background of another, metaphysical time during
  which its intervals are born, age, and recede
  into the past (4) no rate can be assigned this
  passage of things from future to present to
  past.
• III.The futility of attempts to eliminate pure
  becoming from our talk about objects, events, and
  time itself in order to rid ourselves of these
  difficulties.

64
Taylor, Time and Eternity Questions

• 1. Do we really observe or apprehend pure
  becoming (as T. assumes in his description of
  the datum) or do we always observe particular
  changes, qualitative, quantitative, relational,
  coming-into-being and passing-away?
• 2. Explain the cat-and-mouse analogy.
• 3. Granted that its a direct insight that being
  in time is a permanent or structural feature of
  world of our experience, is either the A-series
  (pure becoming present, then past of the future)
  or the B-series (relations of before, after, and
  simultaneity) likewise an immediately intuited
  structural feature of time, or are they rather
  theories about the structure of time?
• 4. Does Taylor privilege any of the three
  dimensions of time (past, present, or future)?
  What would Prior say?

65
Some Time Axioms

• 1. Time is one- (as space is three-) dimensional.
• 2. There is only one (universal) time of which
  all (particular) times are parts.
• 3. Time changes/elapses/flows.
• 4. The flow of time is uni-directional or
  irreversible (flows from the future through the
  present to the past), such that, for example,
  what is done cannot be made undone, i.e. you
  cannot turn the clock back, nor can you
  stop/pause time.
• 5. Relations of before and after are transitive
  and asymmetrical and unalterably permanent.
• 6. Time is impossible without change and change
  impossible without time.
• 7. Time presupposes things existing in time (no
  empty time, pace Kant)
• 8. Nothing can change from future to past without
  having been present
• 9. The flow of time has no rate or speed.

66
Part V Realism and Anti-Realism
67
The Aristotelian Origins of Traditional Realism

• Spoken words are symbola (signs or tokens) of
  pathemata (states) in the soul. And just as
  written words are not the same for all men,
  neither are spoken ones. But the pathemata
  themselves, of which these words are primarily
  semeia, signs, are the same for all, as are also
  the things (pragmata) of which the pathemata are
  likenesses (homoiomata). (Aristotle, De Int.
  16a3-8)
• So
• (1) human thoughts (the states or pathemata of
  the soul) are likenesses, homoiomata, which
  represent (2) pragmata, things, while both (3)
  the spoken (ta en tei phonei) and (4) the written
  word (ta graphomena) (human language) signify
  what is thought about those things. Hereby
  written words again just represent (in the manner
  of signs, semeia) spoken language. Aristotle
  notes that while (3) and (4), the sound of spoken
  and the look of written language, differ for
  different menfor Greeks and barbarians, for
  example(1) the things themselves and (2) the
  states of the soul are the same (tauta) for all
  men

68
Realism Basic Claims

• Truth and falsity are the values an
  epistemically unconstrained bivalent relation
  of correspondence or non-correspondence of a
  belief (thought) or statement (written or
  uttered) to a mind-independently existing object,
  i.e.
• 1. The locus of meaning and truth is the
  judgment
• Beliefs and statements are representations of
  things which must be either true or false
  (bivalence)
• 2.Truth-conditional (correspondence) theory of
  meaning and truth
• The truth of beliefs and statements consists in
  their correspondence with a mind-independent
  world.
• 3. Truth is epistemically unconstrained
• The correspondence (or otherwise) of a belief or
  statement with its object is independent of
  whet