====================
PART I: EXISTENCE
====================


W. V. Quine, _On What There Is_

Pegasus --- Should Pegasus be admitted into an ontology?  Meinong and
followers say yes (and draw a distinction between subsistence and
existence), but Quine says no.  In saying "Pegasus is not" there is no
commitment to a Pegasus object.

possible man --- Quine asks (of Meinongians): Is the possible fat man in the
doorway the same possible man as the possible bald man in the doorway or are
there two possible men?  For Quine, there are no possible men.  (See also
Quine's From a Logical Point of View.)

singular description --- Russell shows how to meaningfully use seeming names
without supposing that there be the entities allegedly named.  "The author
of Waverley" need not have an objective reference to be meaningful. 
"Pegasus" can be rephrased as "the thing that pegasizes".

naming and meaning --- A singular term (such as Pegasus) need not name to be
significant.  There is a gulf between meaning and naming.  There is often a
confusion between alleged named objects (Pegasus) and meanings (the meaning
of the word "Pegasus").  Also, there can be different meanings of the same
named object when named differently (morning star and evening star).  Many
more things have meaning than are names.

commitments --- Only some things we say lead to ontological commitments;
more explicitly, only such things as "There is something (bound variable)
which red houses and sunsets have in common" entail commitments.  To be
assumed as an entity is to be reckoned as the value of a variable; to be is
to be the value of a bound variable.  I think what Quine is saying here (and
I could be wrong) is that there are commitments when and only when there are
sentences of one of two forms: (1) "there exists x such that ..." and (2)
"for all x, where x ranges over a non-null domain, x ...".


*****
Rudolph Carnap, _Empiricism, Semantics, and Ontology_

the debate  --- The problem of abstract entities in connection with
semantics, meaning, and truth. Some semanticists say that certain
expressions including abstract entities like properties/predicates and
propositions/sentences designate certain entities.  Others object strongly
as accepting abstract entities seems to violate the basic principles of
empiricism.

Carnap's aim ---  use of language referring to abstract entities is compatible
with empiricism and does not imply embracing Platonic realism.

external questions --- To speak about a new kind of entities, one needs to
introduce a system of new ways of speaking (framework).  Questions can be
asked within the framework (internal question) and outside of and about the
framework (external questions).  This is a crucial distinction for Carnap. 
External questions are the difficult (in fact, non-cognitive) ones.  Carnap
regards external questions (such as whether numbers are real entities) are
pseudo-questions where those arguing both sides are not using the same
language/background to argue.

semantics --- "To accept the thing world means nothing more than to accept a
certain form of language (framework).... The introduction of the new ways of
speaking does not need any theoretical [external] justification because it
does not imply any assertion of reality."

designata --- "Those who criticize [statements like "The word 'red'
designates a property of things"] do not reject the use of expressions like
"red" nor would they deny that these expressions are meaningful.  But to be
meaningful, they say, is not the same as having a meaning in the sense of an
entity designated."  "Fido" is a name which designates Fido, but "red" does
not designate anything.  (Ryle, hypostatization)

conclusion --- The decisive question for those who use semantical methods is
whether the use of an abstract linguistic form is expedient and fruitful. 
The acceptance or rejection of abstract linguistic forms will be decided by
their efficiency as instruments.  "Let us be cautious in making assertions
and critical in examining them, but tolerant in permitting linguistic
forms."


*****
Bertrand Russell, _Existence and Description_

subject --- general propositions and existence, existence in propositional
calculus, universal and existential quantifiers, names and descriptions.

propositions, prop functions, existence --- Propositions and propositional
functions are often confused but they are very different.  Propositions can
be only true or false (2 values); propositional functions can be necessary,
possible, or impossible (3 values).  (This somewhat parallels true, false,
and defined/exists, as with "Socrates is mortal", "Socrates is immortal",
"there's no Socrates".) Existence is not a property of propositions but only
of propositional functions (to say that unicorns exist is to say that (x is
a unicorn) is possible.  Existence propositions do not say anything about
actual individuals, only about the class or function.

names and descriptions --- Questions of existence concerning names cannot
arise because a name has to name something to be a name.  Often, though,
what appears to be a name is a description; for example, Romulus is "the
person who was called 'Romulus'".  It is a significant statement (whether
true or false) to say "Romulus existed"; it is not trivial because Romulus
(himself) does not appear in the proposition.

definite descriptions --- Definite descriptions are not names; "The
inhabitant of London" is a definite description (although it doesn't refer
to anyone) as is "the sum of 17 and 34".  "The author of Waverley" is a
description, while Scott is a name; English-speakers understand the meaning
of the former, but the latter is not known; "Scott is the author of
Waverley" is not a tautology, as would be the case if both were names.  "The
author of Waverley" is not a constituent of a proposition.

existence --- To say "the author of Waverley exists" (description) is to say
"x writes Waverley" is true for at least one x and at most one x. You can
deny "The author of Waverley is human" by denying that he exists, but you
can't deny "Scott is human" by denying the existence of Scott because Scott
is a name and names require existence.  This is a crucial difference between
descriptions and names.  "The author of Waverley" is an incomplete symbol,
which does not have a meaning by itself, whereas "Scott" is a name referring
to a single object.


*****
Terence Parsons, _Referring to Nonexistent Objects_

purpose --- (1) argue for a difference between the way speakers treat empty
singular terms and the way they treat singular terms that refer to objects
that don't exist; (2) describe a theory about nonexistent objects; (3)
sketch a theory of singular terms.

referring to nonexistent objects --- Referring to nonexistent objects isn't
failing to refer.  Consider the difference between talking about a man in
the doorway that one mistakenly thought they saw and talking about a unicorn
one saw in a dream.  The first case is one of failed reference; here the
speaker realizes their mistaken and treats questions about the properties of
the man in the door as spurious.  The second case is one of reference to a
nonexistent object; here questions about properties about the unicorn seem
meaningful.

paraphrase --- But does this mean there really are nonexistent objects?  Or
is it only that it sounds like there are in ordinary speech but this can be
paraphrased away (as suggested by Carnap, Quine, etc)?  Parsons says that
no one knows how to produce such a paraphrase, and he finds this fairly good
(though not conclusive) evidence for the belief that nonexistent objects
can't be paraphrased away.  So Parsons (like Meinong) thinks there are
nonexistent objects (unicorns, Sherlock Holmes, etc).  (See Meinong's The
Theory of Objects.)

nonexistent object --- What is a nonexistent object?  It's an object that
has a number of properties, but does not have the property of existing.  To
explain this in more detail, Parsons pairs up objects with sets of
properties; for example, the object paired with {goldenness, mountainhood,
...} is a golden mountain, even though there are no real (existing) golden
mountains.  Similarly, there are impossible objects like Meinong's "round
square"; there couldn't be an existing object which is round and square, but
that doesn't prevent there from being an impossible object which is round
and square.  Also, Parsons clarifies a distinction between nuclear
properties (most properties) and extranuclear properties (exists, is
fictional, is possible, is thought about by someone, etc).

singular terms --- singular terms and definite descriptions can be expressed
in a predicate calculus language.  Here quantifiers range over all objects
(not just existing ones).  So we could have "There are winged horses, and
there are no existent winged horses" where the former says "there exists x
where x has the nuclear properties of winged and horse" while the latter
says "there doesn't exist x which has the above nuclear properties and the
extranuclear property of existing".

possible man in the doorway --- There's no such thing as the possible man in
the doorway because there are lots of men there.

truth ---  we can truly say that the fictional detective Holmes was clever,
but we can't truly say that the man in the doorway is clever when there
exists no man in the doorway.

conclusion --- names like "Pegasus" and "Sherlock Holmes" refer; they refer
to nonexistent objects.


*****
W. V. Quine, _Ontological Relativity_

main points --- indeterminacy of meaning, inscrutability of reference,
ontology is (doubly) relative, makes sense only relative to a framework

empiricist --- knowledge, mind, and meaning are part of the same world that
they have to do with, and that they are to be studied empirically.
(antirealist?)

indeterminacy --- learning meaning is a matter of studying the behavior of
others.  There is no "museum" of words containing the fact of the matter
about their meanings.  For naturalism (Dewey, Quine) the question of whether
two expressions have the same meaning has no determinate answer except in
as the answer is settled by people's speech dispositions.

translation between languages --- there is no single correct way to
translate phrases between languages (e.g., "ne ... rien").  There is 
indeterminacy of meaning.

gavagai --- Someone says "gavagai" when pointing at a rabbit, but at the
same time they are pointing at undetached part of a rabbit, temporal stage
of a rabbit, etc.  The only difference between these three is individuation
which Quine says is not something which can be determined by pure ostension. 
There is inscrutability of reference.

within a language --- radical translation begins at home; we use words
(slightly) differently than our neighbors, so there is an inscrutability of
reference here too.

framework --- Reference is not just nonsense; it is relative to our frame of
reference (background language), and relative to that framework we can and
do talk meaningfully; outside of that framework reference is nonsense.

relativistic thesis --- it makes no sense to say what the objects of a
theory are, beyond saying how to reinterpret that theory in another.

circularity --- What makes ontological questions meaningless when taken
absolutely is not universality but circularity.  An answer to "What is an
F?" of "An F is a G" only makes sense relative to the acceptance of "G".

doubly relative --- ontology is doubly relative:  specifying the universe of
a theory makes sense only relative to some background theory, and only
relative to some choice of a manual of translation of the one theory into
the other.


====================
PART II: IDENTITY
====================


Max Black, _The Identity of Indiscernibles_

identity of indiscernibles --- Can there be two distinct objects which share
all the same properties?  If yes, then the identity of indiscernibles is
false.  One character in a dialog tries to present a counterexample with two
indiscernible/identical spheres (Castor and Pollux).  Can two things ever be
truly indiscernible (share all of the same properties)?  But (I think) it
doesn't even make sense to say two things are identical; only one thing can
be identical with itself.

one fallacy --- Here's where I think the fallacy is.  The identity of
indiscernibles (Leibniz's law) doesn't say that if two objects share all the
same properties, they are one and the same object.  This doesn't make sense,
as two objects can't be one object.  The identity of indiscernibles is
(perhaps) actually about names/labels, not about the objects themselves.  It
says if is is appropriate to say that for all properties P P(x) iff P(y),
then x and y are two names for the same object.  Two names can refer to the
same single object, but two objects can't be the same single object. So
there's still the problem of whether two things can have all of the same
properties, but even if two things can have all of the same properties, this
doesn't really disprove Leibniz's law.

another fallacy --- I don't think two things can have all of the same
properties, if we mean some sort of intrinsic properties.  Consider
identical twins.  We might say two twins are identical, but that's really
just shorthand for saying they are identical twins (not identical
simpliciter).  And I don't think it's possible for two spheres to be
identical (the same thing, one object), but it is possible for two spheres
to be identical spheres (or spherewise-identical).  I guess this means I'm
saying that this kind of identity is relative (like Perry?).


*****
Saul Kripke, _Identity and Necessity_

main point --- Statements of identity are metaphysically necessary (not just
contingent).

argument against contingent identity --- (1) The law of the substitutivity of
identicals says that for any objects x and y, if x is identical to y, then
if x has a property F so does y.  (2) Every object is necessarily
self-identical.  (3) Therefore if x and y are identical, they are
necessarily identical.

necessary, a priori --- Kripke says statements are necessary if, whenever
the objects mentioned therein exist, the statement would be true.  A
necessary statement is true and could not have been otherwise.  An a priori
truth is one which can be known to be true independently of all experience.

contingent identity statements --- "The first Postmaster General of the US is
identical with the inventor of bifocals" (Wiggins).  Kripke's answer is
Russell's description; these are descriptions not references.

rigid designator --- A rigid designator is one which designates the same
object in all possible worlds (such as "Nixon" but not "the 37th
president").

Ruth Barcan Marcus --- Whenever a and b are proper names, if a is b, then it
is necessary that a is b.  No contingent identity statements can be made
with proper names.  Kripke agrees with this although he concedes it may seem
false.

Kripke --- If names are rigid designators, then identities are necessary
because "a" and "b" will be rigid designators of a certain thing x, and in
every possible world a and b will both refer to this same object x.  To take
the many examples of "contingent" identity statements as contingent is to
misconstrue the relation between a name and a description used to fix its
reference.

Quine --- Quine argues for contingent identity with the example "Hesperus is
Phosphorus" which seems like an empirical truth which might have come out
otherwise.  Similarly (but not Quine) are claims that other empirical truths
are contingent such as "heat is the motion of molecules", "water is H2O",
"the mental state of being in pain is identical with a brain state".

essentialism --- What are the essential properties of a particular lectern? 
Could this lectern have been made of ice?  Or would it then have been a
different object and not this very lectern?  Butler said everything is what
it is and not another thing; one only has the illusion of contingency.


*****
John Perry, _The Same F_

main point --- Perry argues against Geach's thesis of relative for identity; for
Geach to say "A and B are identical twins" is to say that "A and B are
twin-wise identical" while for Perry (and Frege) it is to say that
"A-the-twin is identical to B-the-twin".  So the question really is, how to
split up statements about sameness: In "A is the same F as B", does F refer
to the way A and B are picked out or to the kind of sameness?

Geach --- Geach's view is that it makes no sense to say "the same" unless we
say "the same F".  When one says "x is identical with y" what is really
meant is "x is the same A as y".  "Being the same horse" does not, for Geach
(but does for Frege) split up into "being a horse" and "being the same as"
(in much the same way as "being a better golfer" can't be split up into
"being better than" and "being a golfer").

Perry/Frege --- According to Perry, the role of general terms is to identify
the referents, not to identify the kind of identity.  x and y cannot be
the same F but different G.

type/token --- Geach thinks a counterexample to Frege is "A is the same word
type as B, but A and B are different word tokens".  However, according to
Perry, Geach is actually referring to two different things when he says "A"
(and "B").  Perry understands the claim as "A-qua-type is the same as
B-qua-type, and A-qua-token is different from B-qua-token" and there is no
counterexample.  

clay/statue --- Similarly, to say "that is the same clay but not the same
statue as you saw last week" is for Geach to say "that object is the
same-clay as before", whereas for Perry it is to say "that statues is the
same as before".  To maintain this, Perry has to say that "this statue" and
"this clay" are not referring to one and the same object; the clay and the
statue are not identical.  This seems like it is very likely to lead to
problems, but Perry thinks that in fact they cannot be the same thing
because they have different (temporal, modal, propositional) properties.

my response --- I actually like both Geach and Perry's views.  I don't think
we are referring to the same thing when we refer to lump-qua-clay and as
lump-qua-statue.  Nor do I think we are talking about the same kind of
identity when we compare two lumps-qua-clay or two lumps-qua-statue.  I
think both references to objects and kinds of identity are relative.  For
example, consider "2 == 2.0" and "2 neq 2.0"; we aren't saying "2 is
numerically-equal to 2.0" (typed operator) nor are we saying that
"2-the-number is equal to 2.0-the-number" (typed variable); instead we are
saying "2-the-number is numerically-equal to 2.0-the-number (both typed
operator and typed variable).


*****
Allan Gibbard, _Contingent Identity_

purpose --- there are contingent identities (non-necessary truths) such as
"The clay statue is identical with the piece of clay of which it is made."

Kripke --- Gibbard takes Kripke as responding to the claim that necessary
truth and a priori truth are the same.  A necessary truth for Kripke is not
a matter of how we can know it but of whether it might have been false if
the world had been different.

clay/statue --- Gibbard concedes that the Hesperus/Phosphorus and other
examples are not instances of contingent identity, but believes he has an
example of a contingent identity as follow: A statue (Goliath) and a piece
of clay of which it is made (Lumpl) are identical if according to their
persistence criteria they have the same duration of existence.

contingent --- Goliath = Lumpl in W1 (this world), but it could have been
the case in W2 (some counterfactual) that Goliath != Lumpl.  Gibbard takes
this to be inconsistent with Kripke's account of proper names.

cross-world identity relative --- what a thing is apart from the way it is
designated is a question without meaning.  Meaningful cross-world identities
of such things as statues must be identities qua something (qua statue or
qua lump).  It makes sense to talk of the same statue in different possible
worlds but no sense to talk of the same thing.  The reference of a name in
the actual world is fixed partly by the persistence criteria.

substitutivity of identicals --- objection to contingent identity is that it
violates Leibniz' Law, for if Goliath is contingently identical with Lumpl,
then although Nec(L=L) and L=G are true, Nec(L=G) is false.  Gibbard's
response is that Leibniz' Law is about properties and relations and doesn't
apply in the relevant context (modal properties, dispositional properties). 
(I basically agree here that Leibniz' Law is true; it is just applied in a
lot of situations where it is not applicable.)

essentialism --- For Gibbard it makes no sense to talk of a concrete thing
as fulfilling a condition in every possible world (necessarily) apart from
designation; it is therefore meaningless to talk of the same concrete thing
in different possible worlds, so there is no worry about essentialism.

de re modality --- Under Gibbard's account there is no such thing as de re
modality for concrete things.

conclusion --- Shown: (1) there is a problem with cross-world identity, (2)
contingent identity is possible, (3) alternative theory of proper names, (4)
while accepting contingent identity and rejecting de re modality for
concrete things, we can still have modal and dispositional properties of
concrete things.

my response --- I agree that if the clay and the statue are identical, that
is a contingent identity, but I don't think the clay and the statue are
identical in the sense of having the same properties (which is the kind of
identity it seems that the other articles have been considering).  They are
identical in having the same matter, taking up the same space, etc.

*****
Stephen Yablo, _Identity, Essence, and Indiscernibility_

point --- There are contingent identities (in fact these are required by
essentialism), but contingent identities are not what Gibbard says they are,
but rather contingent identity is categorical indiscernibility in a world.

essentialism and contingent identiy --- Essentialism both allows for and
requires contingent identity.  

examples --- A bust of Aristotle is necessarily a bust of Aristotle while a
hunk of wax is only accidentally a bust of Aristotle.  Speeding home is
essentially done at a high speed; a certain instance of driving home only
happens to be done at a high speed.

identity-like --- It seems clear that driving is not identical with
speeding, nor the bust with the wax, but that there is some intimate
identity-like connection here.  The driving and speeding have some strong
relation which isn't captured by simply denying their identity.

essence --- An essence is the set of all necessary properties for an entity
to be what it is.  Essences can be built up of other essences; for example,
the essense of the Shroud of Turin is the essence of the Cloth of Turin plus
the property of having enshrouded Jesus.

contingent identity --- If two things agree in all their categorical
properties (in a world), then so far as that world alone is concerned, the
two things are the same; there is a case of contingent identity.  The speeding
home must be done quickly, but the same event when considered as driving home
clearly could have been done slowly; it would then still be driving, but
would not be speeding.

restriction of Leibniz' law --- If a and b are the same concrete thing, then
they have the same categorical properties.


====================
PART III: MODALITIES
====================


Alvin Plantinga, _Modalities: Basic Concepts and Distinctions_

necessity/contingency --- Rather than define necessity and contingency,
Plantinga offers examples: "The average annual rainfall in LA is about 12
inches" is contingent, while "7+5=12" is necessary.

de dicto/de re --- An example of a proposition with modality de dicto is
"Necessarily nine is composite"; an example of modality de re is "Every
animal in this room is necessarily rational."  As an illustration of how
different these are, consider that "What I am thinking of is essentially
prime" might be true while "Necessarily, what I am thinking of is prime" is
false.

essentialism's object to de re modality --- Essentialism here is the claim
that objects have some of their properties essentially or necessarily (and
others accidentally).  This might lead to problems with Leibniz' Law.

Kneale --- Kneale reads sentences of the form "A has F essentially" as short
for "The proposition A has F is necessarily true".

Quine --- Mathematicians may be said to be necessarily rational and not
necessarily bipedal; cyclists necessarily bipedal and not necessarily
rationalist.  What of someone who is both a cyclist and a mathematician? 
(See Quine's Word and Object)


*****
Roderick Chisholm, _Identity through Possible Worlds_

trans-world individuals --- Accepting trans-world individuals (the same
individual in multiple possible worlds) leads to many difficult questions.

identity of indiscernibles --- Suppose Adam in this world lives 930 years
and Adam in another world 931.  In what sense are these the same Adam when
they have a different property?  One way is with world-indexed properties,
which Chisholm shows to parallel time-indexed properties. Just as we can say
"Adam has the property of being-young-at-time-t1 and the property of
being-old-at-time-t2" without contradiction; we can say "Adam has the
property of living-930-years-in-w1" and "Adam has the property of
living-931-years-in-w2".

Adam and Noah --- Imagine a series of possible worlds in which (roughly)
Adam gradually gains properties usually associated with Noah and vice versa. 
Then we end up with a world wN in which Noah-of-wN is indiscernible from
Adam-of-w1 and vice versa.  How do we then tell the difference between w1
and wN?

essential properties --- One way to avoid the Adam/Noah problem is to accept
that individual things have essential properties.  For example, for every
entity x, there are certain properties N and E such that x has N in some
possible worlds (but not all) and x has X in every possible world in which x
exists.  Moreover this are individual essences; for every y, if y has E in
any possible world, then y is identical with x.  That is, the properties E
are essential to x.  (Note a problem with this is that even if there are
essential properties, we have no way of finding out what they are or even
whether there are any.)


*****
David Lewis, _Counterparts or Double Lives?_

purpose --- (1) address problems (such as accidental intrinsics) which
accounts of modality which allow for trans-world identity/individuals must
face; (2) defend counterpart theory against the trans-world individual
theorist's objection (that counterparts do not express real possibility for
the actual individual) by clarifying the counterpart relation; (3) consider
(although I think ultimately reject) individuals as mereological sums of
counterparts as a parallel to 4D perdurance through time.

identity is unproblematic --- Everything is identical to itself and nothing
is ever identical to anything else.  Two things can never be identical; if
they are identical there is only one thing.  We state many problems in terms
of identity, but this is not to say that identity is problematic.

counterparts as representation of trans-world individuals --- It does happen
that something which exists according to one world is identical with
something which exists according to another.  But for Lewis (I think), the
key here is the according to.  It does not happen that something which
exists in one world is identical with something which exists in another.  To
say that Humphrey has possibilities is to say Humphrey is represented in
absentia at other worlds, just as he may be represented in a museum; he is
represented with a counterpart.

objection to counterparts --- Kripke's objection is that counterpart theory
doesn't satisfy our intuition that things could have gone differently for
Humphrey; only that they are different from someone else who is
Humphrey-like.  The claim is that Humphrey could not care less whether
someone else, no matter how much resembling him, would have done otherwise. 
I think the reply is along the lines that this is to misunderstand how
strong the counterpart relation is; it represents possibiilty for the
Humphrey of this world.

trans-world identity --- Humphrey might lead a double life, being in both
the actual and other worlds.  Lewis rejects this.  We need not have Humphrey
himself in multiple worlds to explain possibility.

problem of accidental intrinisics --- if there is overlap between worlds
(the same Humphrey in multiple worlds), how can this common part of two
worlds have different properties in one world than in the other?  Consider:
Humphrey has five fingers but he could have had six (Humphrey has six in
another world).  Lewis rejects having world-indexed properties (Humphrey is
five-fingered-in-w1 and six-fingered-in-w2).  Note that this problem only
arises for accidential intrinsics, not for essential or extrinsic
properties.  However, Lewis' counterpart theory encounters no problem with
accidental intrinsics; counterparts being not some kind of overlap can
differ in intrinsic properties.

persistence and time --- Something persists iff it exists at various points
in time (persists through change).  This can be done through endurance (by
being wholly present at multiple times, 3D) or through perdurance (by having
different temporal parts at different times, 4D).  Lewis accepts the 4D
view.

parallel to time --- Endurance through time is analogous to trans-world
identity of common parts of overlapping worlds; perdurance through time is
analogous to the trans-world identity of a trans-world individual composed
of distinct parts in non-overlapping worlds (no problem of temporal
intrinsics).  Lewis favors something like this composition, and considers
perdurance rather than endurance.

mereological summation and composition --- Consider three cases: I-90 in MA
is I-90 in SD (spatial); the smaller tree in 1980 is the larger tree in
2003 (temporal); Humphrey in w1 is Humphrey in w2 (modal).  None of these
are actually cases on identity; instead what is being said can be rephrased
as: I-90 in MA and I-90 in SD are both parts of the same thing (I-90); the
tree in 1980 and the tree in 2003 are both temporal parts of the 4D tree;
Humphrey in w1 and Humphrey in w2 are both parts of some larger Humphrey
individual who is the sum of all Humphrey-counterparts.  This explains how
counterparts can satisify the intuitition that possibilities exist for
Humphrey (himself).  Lewis considers this to be counterpart theory in
disguise, and in such a disguise that it is more acceptable to those who
usually object to counterpart theory.

problems with the above analogy --- The parallel between temporal and modal
composition breaks down.  Temporal parts of an ordinary thing are related by
causal dependence; modal parts are not.  Temporal parts have an ordering;
modal parts do not.  A person perduring through time has a collective
self-interest; a person as the sum of counterparts does not.

vagueness --- The reason it's vague where the outback begins is not that
there's this thing, the outback, with imprecise borders; rather, there are
many things, with different borders, and we don't enforce a choice of one of
them as the official referent of the word "outback".  Vagueness is semantic
indecision.


*****
Robert Adams, _Primitive Thisness and Primitive Identity_

purpose --- Adams defends his acceptance (non-qualitative) primitive thisness.

thisness and suchness --- A thisness (haecceity) is the property of being
identical with a certain particular individual.  This is similar to an
essence except it is nonqualitiative.  A suchness is a qualitative property
(example given is being a home-owner).  For Leibniz, thisnesses are
suchnesses; that is, there are individual essences which are qualitative
properties which pick out individuals.  Adams understands thisnesses as a
primitive and not reducible to suchnesses.  I think this means that Adams
rejects the bundle theory of particulars.

against the identity of indiscernibles --- I think Adams holds that it is
possible for two things (like Black's spheres) to be qualitatively
indiscernible but nonqualitatively discernible (by having different
thisnesses) so that qualitative indiscernibility does not imply identity.

almost indiscernible --- Adams provides an example of two near-indiscernible
twins with only one qualitative difference.

transworld identities --- Adams believes that accepting nonqualitative
thisnesses leads to a strong argument for primitive trans-world identities.


*****
D. M. Armstrong, _The Nature of Possibility_

purpose --- Armstrong defends a naturalistic, combinatorial theory of
possibility.  (That is, all there is the space-time world and possibilities
are states of affairs of combinations of individuals and properties.)

ontology --- Armstrong accepts atomic states of affairs, facts, individuals,
monadic properties, and relations.

states of affairs --- Atomic states of affairs are things like "a's being F"
and "b's having R to c".  Armstrong doesn't accept disjunctions and
negations of atomic states of affairs are not allowed, but he does accept
conjunctions to build up molecular states of affairs.  There are also
possible and merely possible states of affairs (and impossible states of
affairs).  Merely possible states of affairs don't exist or have any sort of
being, but we can refer to them.  Simple individuals, properties, and
relations may be combined in all ways to yield possible atomic states of
affairs.

Wittgenstein worlds --- Molecular states of affairs thought of as totalities
of being are possible worlds.  A possible world must contain each simple
individual, property, and relation, and nothing else.

alien universals and individuals --- Armstrong denies the possibility of
genuninely alien universals.  The possible is determined by the actual.  (I
think the example he gives is that if there's no such actual universal
redness, there's no possible universal redness; universals must be all
possible states of affairs if they are to be in any.  The combinatorialism
is an actual world chauvinist.  Alien individuals seem possible.  Also note
that Armstrong (roughly) says that quidditism is incompatible with alien
universals and haecceitism is incompatible with alient individuals.

contraction --- Any universal is contingent and need not be in all possible
states of affairs.  States of affairs must contain something and may not
contain more than is in the actual world, but they can contain less than is
in the actual.


====================
PART IV: UNIVERSALS
====================


D. M. Armstrong, _Universals as Attributes_

purpose --- To develop an ontology which allows for universals, but only
instantiated universals, and only some of those (sparse Aristotelian
realism), to add states of affairs to the ontology, and to consider thin and
thick particulars and the relation between particulars and their properties.

realism/nominalism --- For realists there are universals; Platonic realists
hold that there are uninstantiated universals (universals before things)
while Aristotelian realists hold that there are only instantiated universals
(universals in things).  Nominalists deny that there are universals
(universals after things); nominalists include predicate, concept, class,
and resemblance nominalists.  Armstrong is an Aristotelian realist. 
Properties (universals) are the constituents of things.

uninstatiated universals --- Armstrong rejects arguments for uninstantiated
universals, such as the argument from meaning (that talking about unicorns
is meaningful so there must be a corresponding universal for there to be
meaning), and the idea of universals as ideals that are approximated.

restricting universals --- Armstrong rejects disjunctions and negations of
properties as universals, accepting only simple properties (including
relations) and conjunctions.  (I think he also rejects "being identical" and
others.) There is no automatic passage from predicates (linguistic entities)
to universals; here Armstrong takes himself to be agreeing with a main point
of Wittgenstein's families of games example.  While it's clear to Armstrong
that not everything is a universal, he doesn't think there's an infalliable
way of deciding what are the true universals, and if they are to be
determined it is through science.

states of affairs --- In addition to accepting particulars and universals
(the properties which particulars instantiate), it seems that states of
affairs (what others may call facts) are needed to tie them together (a's
being F).  Reasons for accepting states of affairs include that states of
affairs are plausible candidates for the terms of causal relations and that
states of affairs can explain multiple location of property universals. 
Note that states of affairs are required by those who recognize properties
and relations, not just as universals but also as particulars (tropes). 
Think of the world as a world of states of affairs, with particulars and
universals having existence only within states of affairs.

particulars (thin and thick) --- Particulars outside of states of affairs
are bare particulars, not clothed in any properties or relations.  When we
say "a is F" that's the is of instantiation, the fundamental tie between
particulars and properties.  Thin particulars are particulars apart from
their properties, linked to properties by instantiation but not identical
with them.  Thick particulars are particulars involving their properties;
thick particulars are states of affairs (enfolds both thin particulars and
properties held together by instantiation).

multiple instantiation --- A common charge agains Aristotelian realists is
How can universals be located in two places at once?  Armstrong's reply is
that once properties and relations are thought of not as things but as ways
it doesn't make sense to think of them as floating around anywhere.  Talk of
locating universals in space-time is a crude way of speaking.  Universals
are constituents of states of affairs; universals are in space-time by
helping to constitute it.

*****
David Lewis, _New Work for a Theory of Universals_

purpose --- Lewis rejects to Armstrong's theory of universals, concluding
that (not-sparse) properties are needed for some purposes and that
universal-like entities (natural properties) are useful in many ways
(duplication, supervenience, divergence, materialism, laws, causation,
language and thought).

possibilia --- Although Lewis is a nominalist, he admits possibilia,
particular individual things some of which compromise our actual world and
others of which are unactualized.  Lewis understands properties as classes
of possibilia.

universals vs properties --- Lewis takes universals to refer to Armstrong's
(sparse and instantiated) universals.  In contrast, Lewis' properties are
classes; to have a property is to be a member of the related class. 
Universals are wholly present where instantiated; properties are spread
around (particulars are members of properties).  Universals are sparse;
properties are abundant (any class of things is a property).  This abundance
of properties is a feature for some explanations (semantic predication,
intensional attitudes), but doesn't help to explain others.  Lewis seems to
be saying that universals may be needed, but so are properties.

natural properties --- Natural properties are a (spare) subset of
proprerties, posited to explain resemblance and causal powers (among other
things).  Lewis' natural properties correspond to Armstrong's (intrinsic)
universals.  As such Lewis believes universals are not needed and replaces
universals with natural properties; that is, Lewis' ontology contains all
properties and marks out an elite subset of them as natural properties.

uses of natural properties --- Uses of natural properties include:
duplication (two things are qualitative duplicates if they have exactly the
same perfectly natural properties); supervenience/reduction (where "A
supervenes on B" means any difference in A requires a difference in B);
divergence of worlds (two worlds are divergent if they have duplicate
initial temporal segments, contrasted with branching); materialism (that
physics is a correct, complete, comprehensive theory of the world); laws and
causation (there is no necessary connection, unlike for Armstrong, and
causation holds between events); and language.  (Note that this summary is
quite skimpy on the details; refer to Lewis's text for details.)


*****
Sydney Shoemaker, _Causality and Properties_

purpose --- Shoemaker explain his theory that properties are causal powers
and that causal relations hold between particular events in virtue of the
properties possessed by the constituent objects of those events.

events --- Events are usually taken as the terms of the causal
relationships, but events consist of objects and their properties.  That one
event causes a second is explained by the constituent objects of the cause
event.

geunine properties --- A property is geunine iff its acquisition/loss by a
thing constitutes geunine change.  This is distinguished from mere Cambridge
changes/properties such as being grue or being 17 miles south of some object
(Geach).  Geunine properties are the ones Shoemaker thinks are relevant for
an explanation of causality.

Hume --- Shoemaker, while holding a position which is in some ways similar
to Hume's, is not a Humean. 

powers and properties --- For something to have a power is for it to be such
that its presence in circumstances of a particular sort will have certain
effects.  It is in virtue of a thing's properties that the thing has the
powers that it has (parallels a distinction made by Locke).  Powers can be
thought of as functions from circumstances to causal effects; properties are
in some sense second-order powers (powers to produce first-order powers if
combined with other properties).  What makes a property the property it is
is its potential for contributing to the causal powers of the things that
have it.

conditional power --- When a thing has a power conditionally upon the
possession of certain properties, this amounts to its having a conditional
power.  (Example: A knife-shaped object has the conditional power of being
able to cut wood if knife-sized and made of steel).

epistemological reasons --- Shoemaker holds his theory of properties because
he believes that only if some causal theory of properties is true can it be
explained how properties are capable of engaging our knowledge/language in
the way they do.  We know/recognize properties by their effects.

theoretical simplicity --- If the water supply was poisonous all day long,
it is more plausible to suppose this was due to one posionous substance
rather than several.

identity across time/worlds --- There is a close link between identity
across time and identity across possible worlds.  The power or possibility
of causing something must be present at time t if that something is to occur
at time t+1.

essences --- The totality of a property's causal potentialities constitutes
an individual essence.

necessity --- For Shoemaker, causal necessity is a species of logical
necessity.  Following Kripke, necessary truths need not be a priori. 
Shoemaker defends the sort of "necessary connection" which Hume refuses.