Part IB, Metaphysics & Epistemology Dr Will Davies (wd243)

Mon 2pm, LB1

PRIMARY AND SECONDARY QUALITIES

WEEK 1
1. Overview

Week 1: Introduction to the Primary/Secondary Quality Distinction.

Week 2: Dispositionalist Views of Colour

Week 3: Colour Physicalism

Week 4: Colour Primitivism

2. Key Questions

1. What distinguishes primary qualities (PQs) from secondary qualities (SQs)?

2. Which properties, if any, are secondary qualities?

 We’ll focus on colours as a test case, as they are typically offered as paradigmatic
examples of SQs.

3. Locke on the Primary/Secondary Quality Distinction

During Locke’s time at Oxford, he was mentored by the scientist and theorist Robert
Boyle, from whom he acquired a mechanical, atomistic, corpuscularian view of physics.
This view holds that the fundamental or basic units of the world are indivisible particles
or atoms, which have properties such as solidity, shape, and extension. Atoms are
contained within the void or space, and are either in motion or at rest. The most relevant
contrast here was with Cartesian plenum theorists, who denied the existence of the void,
and held that all matter was infinitely divisible. Within this atomistic worldview, there
seemed to be no place for sensible qualities such as colours, smells, tastes, sounds, and
so on.

The PQ/SQ distinction was intended to help accommodate these sensible qualities
within an atomistic, mechanistic, account of matter.

1
Part IB, Metaphysics & Epistemology Dr Will Davies (wd243)
Mon 2pm, LB1

The Problem of the Manifest

Although Locke’s version of the PQ/SQ distinction was motivated by his commitment to
17th Century atomistic view of physics, the distinction has its origins in a much older
and broader philosophical problem: The Problem of the Manifest. Sellars (1962) takes
this to be one of the central and defining problems in the history of philosophy:

There seems to be a significant disparity between the view of the world provided
by our best physical scientific theories – the scientific image – and the view of the
world provided by ordinary sense perception, and our thought and talk about
objects in the environment – the manifest image. This presents a classic location
problem: where are the colours, smells, tastes, and sounds of ordinary perception
situated? How are we to reconcile the manifest image with the scientific image of
the world?

This problem has ancient origins. Both Locke’s atomistic physics and the problem of the
manifest are foreshadowed in Democritus’ remark, ‘by convention hot, by convention
cold, but in reality atoms and void’. As Smith (1990: 224) notes, through the 17th
Century, ‘figures such as Descartes and Boyle, Leibniz and Newton, while holding
markedly different physical theories, can be in agreement over the general character of
the primary-secondary quality distinction.’ The distinction stems from the thought, as
Stroud (2002: 7) puts it, that ‘the world presented to human beings in their daily lives
diverges from, and appears to be a much richer place than, that austere reality
described by the atomists.’

Characterising Primary Qualities:

According to Smith (1990: 233), Locke held that ‘it is necessary and sufficient for a
quality to be primary that it characterise everything material: that which attaches to a
material entity simply in virtue of its being material.’ In Locke’s terminology, this is to
say that PQs are Original. PQs were intended to provide metaphysical unification, being
qualities that are present and quantifiable/measurable in all matter, and hence apt to
feature in systematic scientific explanation. This served the central aim of enlightened
science, which was to generate systematic explanations of physical phenomena, rather
than piecemeal explanations linking certain types of cause with certain types of effect.
See Smith (1990: 225-226).

2
Part IB, Metaphysics & Epistemology Dr Will Davies (wd243)
Mon 2pm, LB1

Characterising Secondary Qualities (following Bennett 2001, Ch. 5 on Locke):

1. SQs are dispositions to cause characteristic types of sensory states. Locke: ‘Secondary
qualities . . . are nothing but the powers those substances have to produce several ideas
in us by our senses’ (II.xxiii.9)

 Bennett (2001) argues that 1 is the most central and important thesis about SQs.
This view has gained widespread currency and provides the default
interpretation of ‘secondary quality’ in the contemporary literature.

2. SQs are not in outer objects. Locke: ‘Yellowness is not actually in gold’ (xxiii.10).

 This seems to be inconsistent with 1, which holds that SQs are powers in outer
objects to cause certain sensory experiences. Compare the disposition to cause a
spark, which is definitely a power in or of the match.

3. SQs are not intrinsic to objects, but are relations between an object and something
else. Locke: The yellowness, solubility, etc. of gold ‘are nothing else but so many
relations to other substances, and are not really in the gold considered barely in itself’
(II.xxiii.37).

 The fact that SQs are individuated with respect to causal relations to perceiving
subjects does not entail that SQs are themselves relations. Compare being the
brother of ___, which is clearly relational, with being a footprint, which is a
monadic property that is individuated w.r.t. causal relation to a foot.

4. SQs are properties of minds/experiences rather than objects. Locke: ‘Light, heat,
whiteness or coldness are no more really in them than sickness or pain is in manna.
Take away the sensation of them; let not the eyes see light, or colours, nor the ears hear
sounds; let the palate not taste, nor the nose smell; and all colours, tastes, odours, and
sounds, as they are such particular ideas, vanish and cease’ (viii.17).

 This seems to involve a slip from SQ view, which takes colours to be properties of
ordinary objects, to Galilean subjectivism, which holds that colours are properties
of experiences that are mistakenly projected onto ordinary objects. SQ view of
colour is realist: it preserves the truth of the ordinary claim that redness is a

3
Part IB, Metaphysics & Epistemology Dr Will Davies (wd243)
Mon 2pm, LB1

feature of the tomato. In contrast, Galilean subjectivism is irrealist: it commits us

to global error in mistakenly attributing colour properties to ordinary objects.
 N.B. There is significant debate as to whether Locke himself was a SQ theorist in
the sense of 1. For example, Smith (1990: 236), following Alexander (1976-77),
argues that Locke was in fact a Galilean subjectivist about colours. Whatever the
truth of the matter, this does not undermine the mainstream understanding of
SQs as characterised by 1.

4. Some Confusions to Avoid

1. It is not true by definition that colours, tastes, smells, and sounds, are SQs. One could
maintain the conceptual distinction between PQs and SQs while insisting that colours
are in fact PQs, as we’ll see in Lecture III.

2. PQ/SQ distinction is not the same as the distinction between categorical and
dispositional properties. Although all SQs are by definition dispositions, not all
dispositions are SQs. For example, brittleness is a disposition but not a SQ. (This would
class as a tertiary quality on Locke’s view.)

3. PQ/SQ distinction is not the same as the distinction between ‘objective’ and
‘subjective’ properties. For example, the property of being an experience as of red is
arguably ‘subjective’, but it is not a SQ.

4. PQ/SQ distinction is not the same as the distinction between ‘physical’ and ‘non-
physical’ properties: the existence of SQs does not entail the falsity of physicalism. E.g.
one could hold that SQs supervene on microphysical properties.

5. The PQ/SQ distinction is not exhaustive. Besides tertiary properties, such as fire’s
power to melt wax, there are numerous other properties that are not apt to feature in
systematic, fundamental, physical explanation. E.g. the property of being a hammer.

5. Is the Primary/Secondary Quality Distinction Satisfactory?

Problem: SQs are claimed to be dispositions to produce certain types of sensory

experience. Any square thing will be disposed to appear square to normal subjects, but
squareness is supposed to be a paradigmatic PQ. Doesn’t this collapse the distinction?

4
Part IB, Metaphysics & Epistemology Dr Will Davies (wd243)
Mon 2pm, LB1

Bennett’s (2001) Response: Possessing the property of squareness bestows a range of

other causal powers besides the power to affect perceivers. In contrast, the one (only?)
significant causal effect of being red is appearing red to normal perceivers. SQs such as
colours are thus distinguished from PQs such as shapes by the extent to which they are
causally irrelevant or inert.

Another Response: We need to distinguish various different theses relating the property
of squareness to sensory experiences:

(DISP) If x is square, then x is disposed to produce characteristic experiences of

type S.

(IDENTITY) The property of squareness=the disposition to produce

characteristic experiences of type S.

(BICON) Necessarily, x is square iff x is disposed to produce characteristic

experiences of type S.

Everyone can (and should) accept DISP: all square things have the disposition to appear
square. But this doesn’t imply that squareness is a SQ. SQ theorists would have to be
committed to IDENTITY or BICON. The SQ view is supposed to tell us something about
the nature or essence of these properties, not simply materially necessary conditions.

Readings:
P. Alexander, "The Names of Secondary Qualities," Proceedings of the Aristotelian Society
78 (1976-1977).
J. Bennett, Learning from Six Philosophers Vol. 2 (Oxford: Clarendon, 2001), Ch. 5.
J. Locke, Essay Concerning Human Understanding, Book 2, Ch. 8.
W. Sellars, “Philosophy and the Scientific Image of Man,” reprinted in Science, Perception
and Reality, Routledge & Kegan Paul Ltd; London 369–408.
A. Smith, “Of Primary and Secondary Qualities”, Philosophical Review, 99 (1990):221-54.
B. Stroud, The Quest for Reality: Subjectivism and the Metaphysics of Colour (Oxford:
Oxford University Press, 2000).