Truth and The World: DR David J. J. Austin University of York Spring Term 2023 (PHI00137H)
Truth and The World: DR David J. J. Austin University of York Spring Term 2023 (PHI00137H)
Truth and The World: DR David J. J. Austin University of York Spring Term 2023 (PHI00137H)
Austin
University of York
WORLD Lecture 6
RECAP
TRUTH-MAKER THEORY
Truth-maker theory says that there is an intimate link between truth and
ontology:
Between what is the case and what there is.
According to truth-maker theory, for a proposition to be true requires there
to be some entity or entities that make it true.
The truth-makers are the ontological ground of the truth – their existence
explains why the proposition in question is true.
In both the lecture and seminar last week (taking propositions as the
primary truthbearers), we looked at the following question:
What exactly is the nature of the truth-making relation, i.e. between a
true proposition and its truth-maker?
To ask this question is to probe what sort of analysis, if any, can be given
of the truth-making relation.
(1) Propositions are abstract entities, and thus simply not the kind of thing
that can enter into causal relationships.
(2) Causal connections are contingent connections, in the sense that the
entities related by cause and effect might both have existed but not been so
related—my kicking of the ball caused the breaking of the window, but
both might have existed and not been so related, had you caught the ball
but something else broken the window.
Properties are those entities that can be predicated of things or, in other
words, attributed to them. Properties are often called predicables.
Properties are also ways things are, entities that things instantiate.
For example:
If we say that this is a leaf and is green, we are attributing the properties
leaf and green to it, and, if the predication is veridical, the thing in question
exemplifies these properties.
Although we will get to the sort of ontology we might call upon to ensure
that our discourse about properties is true, the question we’re asking now
is:
What sorts of entity were being referred to by the subject term in the
sentence (i.e., (1)-(4))?
It would be natural to think that the term ‘the table’ is picking something
out and so the referent of the term is a part of the truth-maker for the
propositions expressed by each of (1)–(4).
TRUTH AND THE WORLD 2023 29
THINGS
Let’s begin with the observation that it appears that there is something
common to each of (1)–(4):
Something that is picked out by the term ‘the table’.
Each of them seems to ascribe some properties to the table.
The natural thought, then, is that the table itself is distinct from those
properties.
There is a table and the table has all of the properties described above.
TRUTH AND THE WORLD 2023 30
THINGS
P1.We think that ordinary objects (tables, horses, stones etc.) have
properties.
P2.It seems to us as if these properties are united in some way.
C: This suggests that there is something that plays the role of uniting the
disparate properties. This is substance.
Originally, it was suggested that the obvious inference was that because
there appeared some thing in common in each of (1)–(4), that was not
identical to the properties, we should posit a thing – substance.
That thing, however, has caused a problem, so let’s just drop it and say the
following instead:
Objects are just bundles of properties.
Transcendent Universals:
• For a to be F is for a to stand in relation R (instantiation) to a
transcendent universal.
• ‘Transcendent’ because there’s a realm of universals = non-spatio-
temporal location. They are abstract objects.
• The view is Platonic. We may even say ‘the Form of F’
Transcendent Universals:
Particular a does not just instantiate part of the ‘form’ F, because different
particulars would then instantiate different parts of F.
We would then lose the solution to the problem of universals: we would
have the problem of explaining why two F things are similar just in virtue
of instantiating different parts of the Form. After all, I have different parts,
but they’re not all similar.
So, R must be a relation between a and the whole of F.
TRUTH AND THE WORLD 2023 52
PROPERTIES AS UNIVERSALS
Transcendent Universals:
The view commits us to properties as existing outside space and time and,
thus, to abstracta.
And as we know, there are substantial concerns with positing abstract
entities.
So, we had better have some good reason for thinking that properties exist
outside space and time.
Transcendent Universals:
Note that Edinburgh is to the north of London.
Notice, also, that ‘to the north of’ seems to be something that is repeatable.
Edinburgh is to the north of London, as are Sheffield, Nottingham, Leeds,
Durham and Peterhead.
We might think, therefore, that there is a universal ‘north of’.
Russell’s argument, then, is that if the relation exists, but is in neither space
nor time, then the universal must exist quite apart from its instances that
appear in the world!
But here’s a thought…
It seems to make perfect sense to say of properties and relations that they
exist whenever their instances do.
That is, although there is no sense in which the relational property ‘to the
north of’ exists at some place and time to the exclusion of all others, it is
nonetheless something that exists in space and time by virtue of ‘being to the
north of’ being something that is instantiated by concrete particulars –
Edinburgh and London.
This leads us, then, to the view that properties are immanent universals.
Immanent Universals:
• In contrast to transcendent universals, the view of universals as
immanent treats them as existing only at their instances.
• Were it to be the case that no red objects exist, then the colour red would
not exist.
• This view is broadly Aristotelian.
Immanent Universals:
It is preferable to avoid positing universals as distinct from their instances,
in addition to their existing at their instances – and concerns about what it
actually means to exist ‘outside space and time’.
It becomes apparent that we have substantial reason to prefer an
Aristotelian view of universals – that they are immanent.
An important distinction:
• Sparse vs. abundant properties
Be aware that not every predicate seems to denote a property. Consider the
following sentence:
The sky is not green.
This sentence is true and ‘not green’ seems to perform as a predicate.
It seems to make sense, then, to ask whether we need a further universal of
‘not green’ that the sky need instantiate?
After all, we have said that predicates denote properties.
TRUTH AND THE WORLD 2023 61
SPARSE VS. ABUNDANT
PROPERTIES
It might be thought, however, that it is more natural to think that the truth-
maker for the sentence, ‘The sky is not green’, just is the sky’s being blue,
rather than the sky’s instantiating some further property ‘being not green’.
It would be natural to think that sentences like this show us, not that there
are properties such as ‘being not green’, but that not every predicate
denotes a universal.
Rephrase ‘The sky is not green’ as ‘it’s not the case that the sky’s green’.
One very intuitive thought is that we should restrict our account of which
predicates denote properties to only those properties that:
(a) make a causal difference or,
(b) those that make a difference to what is true.
After all, if a property does not make a causal difference, and does not
make a difference to what is true, then it is hard to see why we should want
to posit it.
To see these principles in action, consider the case we might make against
disjunctive properties – that is, properties that correspond in their structure
to instances of disjunctive predication:
(D) x is red or blue.
If x has is red, then (D) is true (that’s inimical to the nature of a disjunction
– it’s true if either disjunct is true).
So, do we then need to think of x as instantiating a disjunctive property?
Notice further that if x instantiates the property red, then x and red make
true <x is red or blue>.
Given that this is so, we can preserve the truth of (D) without recourse to
including any disjunctive universals in our ontology.
Similar considerations as these can obviously be brought against ‘negative’
properties, too. If an object has the property blue, like the sky, then no
further causal or truth-making work would be done by it also positing the
property of ‘not green’, like with ‘the sky is not green’.
TRUTH AND THE WORLD 2023 70
SPARSE VS. ABUNDANT
PROPERTIES
One way to make sense of this is to make a distinction between sparse and
abundant properties:
Abundant properties are such that there is a property for every descriptive
condition that you care to give. Our disjunctive properties could go on this
list, as could nearly any other predicate that you specify.
So, there is an abundant property of ‘being a banana or a cat’, a property of
‘being square and yellow’, and so on. But these predicates are not
the interesting predicates.
TRUTH AND THE WORLD 2023 71
SPARSE VS. ABUNDANT
PROPERTIES
One way to make sense of all this is to make a distinction between sparse
and abundant properties:
Sparse properties will be relatively few – only those properties that are
posited by our ultimate and best physical theory. These properties – the
ones uncovered by our best physics – may also be called ‘natural’.
For example:
The chair you’re sitting on satisfies the predicate ‘is black’, and so there is
a property of being black.
But the property is not perfectly natural.
Instead, ‘the chair is black’ is to be understood as made true by a particular
construction of natural properties:
E.g., properties of mass, charge, spin, etc.
TRUTH AND THE WORLD 2023 75
SAMPLE FOOTER TEXT
LET’S TAKE STOCK
What makes (OT) true? Assuming bundle theory (though not committing
ourselves to any particular view of properties for the time being), we’ll
need:
(1) a bundle of properties that is to be identified as the table; (2) a
further bundle of properties that is the desk; and (3) a relation – to the
north of – that stands between the two bundles.
This collection of entities, or so it would seem, is the truth-maker for (OT).
The list of entities (the office desk, the relation ‘to the north of’ and the
office door) makes no mention of the ways in which the entities on the list
are arranged with respect to one another.
Indeed, these three entities could make true (OT*):
(OT*) The office desk is to the north of the office door.
It will all come down to how we arrange them with respect to one another.
This is a problem:
What makes it true that the office door is to the north of the office
desk can’t also make it true that the office desk is to the north of the
office door!
Thus, sentences (OT) and (OT*), though similar, will clearly require
different truth-makers .
Suppose that we allow ‘R’ to stand for any relational term (such as ‘to the
north of’) and that we let ‘a’ and ‘b’ stand for any two particulars.
What we want is to have different truth-makers for the expressions Rab and
Rba.
At the moment, however, we seem to be lacking that. All we have as truth-
makers are the particulars, a and b, and the relation R.
We need something more…
TRUTH AND THE WORLD 2023 84
Armstrong makes the point, here:
‘Let a love b, and b love a. The two
states of affairs are presumably
independent. Either could have occurred
without the other […] Yet the two
different states of affairs involve exactly
the same constituents. How are they to
be differentiated?’
Importantly, for Armstrong, there are (at least) two different senses of ‘part’
available to us:
(1) Standard mereological parthood, where the fusion of any particulars a, b
and c, is identical to the fusion of b, c and a.
(2) ‘Non-mereological parthood’.
But why does Armstrong introduce the second sense?
Suppose, however, that we have a state of affairs, S, which has as its non-
mereological parts, the relation R, and the particulars a and b, and that they
form the state of affairs, Rab.
The state of affairs S is not identical to the state of affairs, Q, that includes
the same constituents, but that is non-mereologically composed such that
Rba.
Armstrong does not say that much about what, exactly, non-mereological
composition is.
It is, in some senses, a primitive notion in his theory.