Banner-ohne Ort+Datum

Programme and Abstracts

Home Programme and Abstracts | About us | Accessibility


WEDNESDAY, 26 February 2020

16.00 c.t. – 18.00 pm     Keynote address by MICHAEL AYERS
Primary and Secondary Knowledge, Evidence and Defeasibility


10.00 am – 10.30 am     Symposium opening session

10.30 am – 12.30 pm     Session 1 – Knowledge and metaphysical commitments

TOM CROWTHER: Substances, events and processes

12.30 pm – 14.30 pm    Lunch break

14.30 pm – 17.30 pm    Session 2 – Knowledge and perspicuity

MARIA ROSA ANTOGNAZZA: Knowledge and the First Person
GUY LONGWORTH: Knowing, Knowing Perspicuously, and Knowing How One Knows
RORY MADDEN: Perspicuity, Reflexivity, and Consciousness

18.30 pm    Conference dinner


10.00 am – 12.30 pm    Session 3 – Knowledge and perception

CHARLES TRAVIS: In the thick of it

12.30 pm – 14.15 pm    Lunch break

14.15 pm – 15.00 pm    Vienna Circle tour guided by Christoph Limbeck-Lilienau

15.00 pm – 17.00 pm    Session 4 – Knowledge and scepticism

ARNAUD DE COSTER and PAUL TUCEK: Ayers on Scepticism, Certainty and Defeasibility


Michael Ayers: Primary and Secondary Knowledge, Evidence and Defeasibility


I start (as requested) with a resumé of the argument of Knowing and Seeing on the following lines, both for the sake of anyone present who hasn’t read the book and in order to emphasise the intended unity of its argument. I will then focus on the discussion of ‘defeasibility’, and perhaps some other points at which it occurs to me that the argument might profitably be restructured. I expect to end with some general reflections on some of the troublesome peculiarities of the notion of knowledge.

Knowing and Seeing

In Part I, Chapters 1, 2 and 4 each take a different approach to a key distinction drawn between primary and secondary knowledge: historical, phenomenological and linguistic respectively.

Chapter 1 documents, explains, and to an extent defends the sharp distinction between knowledge and belief that shaped epistemological theory from Plato to Locke. ‘Knowledge’ is grounded in the immediate apprehension of (‘seeing’ or ‘perceiving’ by the senses or by the eye   of reason) reality or intelligible truth. ‘Belief’, in contrast, is at best grounded on ‘extraneous reasons’ (we may believe truly what a mathematician tells us, but only know it if we ‘see why’ it is true). We have to do our own knowing. What is known is simply evident, whereas justified belief draws on evidence for it. ‘Knowledge’ comes with knowing that we know. A mistake made by   Plato, that was often later repeated, was to suppose that ‘knowledge’ is therefore the product of an infallible faculty.

This traditional, technical distinction between ‘know’ and ‘believe’, involving the narrowing of the everyday senses of both (which allow that what we know, we believe) became unpopular with the influence of Hume’s naturalistic scepticism. That is unfortunate, since it points to the crucial distinction between primary and secondary knowledge that is explained in Chapter 2. In that chapter, a close analytic phenomenology of sense perception, with an eye on current empirical psychology of perception, both endorses direct realism and explains why it is true that sense perceptual knowledge comes with knowledge that and how we have it. The argument appeals to the integration of the senses (currently an object of intense empirical research) and its connection with the point that in perceiving any object, at least normally one is ipso facto perceptually aware of its causal-cum-cognitive relation to oneself. As the metaphor of ‘seeing’ for understanding might suggest, perceptual knowledge is the paradigm of ‘primary knowledge’– i.e., direct and perspicuous knowledge. Moreover, in place of the assumption of much recent analytic theory of knowledge that an ‘analysis’ or definition covering every case of belief that we would ordinarily count as knowledge is what is required in order to explain what knowledge is, it is argued that whether a belief of any kind properly counts as knowledge, whether primary or secondary, is determined by the extent to which it is, in one way or another, analogous to the paradigm. It is further argued that no being could have secondary knowledge, or indeed mere belief, without some primary knowledge.

The next chapter, Chapter 3, starts with a very brief history of ‘conceptualism’, moving on to consider critically the different explanation of perceptual knowledge and its authority, and of why it involves knowing that one knows, that is offered by a leading, currently active conceptualist, John McDowell. A main bulwark of analytic conceptualism, the ‘sortalist’ approach to the identity of physical objects at a time and over time (as expounded, e.g., by David Wiggins) is criticised in the light of the conclusions of Chapter II and of fundamental differences in the classification and individuation of substances on the one hand and of ‘modes’ (such as events, properties and states of affairs) on the other, differences palpable in the deep structure of natural languages.

Chapter 4 discusses in some detail the ‘ordinary’ (or ‘natural’) language used to speak about perception and knowledge, starting with an examination of so-called ‘embedded’ sentences and comparison of their functions in ascriptions of belief, in ascriptions of knowledge, and after verbs of perception. Can it be said that we believe propositions but know facts, and what are facts anyway? A question considered is the relation between the ‘evidence’ of what is evident and the evidence for a conclusion, and why the latter can be a statement, a fact, an event, a state of affairs or a physical object such as a bottle of arsenic? Similarly, what is the relation between knowing that S is P, knowing S and knowing of S (‘propositional knowledge’, ‘knowledge by acquaintance’ and ‘knowledge by description’)? In particular, is it the case, as many epistemologists assume (e.g., McDowell, Tim Williamson, Fred Dretske and others discussed in Part II), that propositional knowledge has priority, constituting the fundamental subject matter of epistemology? (Answer: No, it isn’t.)

Chapter 5 subjects scepticism (in particular, versions of the ‘Cartesian’ argument) to detailed critical examination in the light of the conclusions of Part I, moving on to target its uncritical ‘methodological’ employment in recent and current epistemology. Explanations are given of the relation between knowledge, probability and certainty, of the precise way in which certainty (and so the ascription of knowledge) is relative to evidence, and of the senses of ‘fallibility’ and ‘defeasibility’ that are useful in epistemology. Briefly and broadly, all our cognitive faculties are, like the senses, fallible, and their deliverances are defeasible; that is: their deliverances have authority (as explained in Chapter 2 with respect to the senses) and in general deliver knowledge, but can be ‘defeated’ by ‘special reasons’. The authority means that the burden of proof lies on the doubter or denier, analogously to the burden of proof in a legal case, given a defeasible or ‘rebuttable’ principle (e.g. the principle accepted in British law that intentional homicide is murder). What then are ‘special reasons’ such as can be brought against a claim to knowledge?

Ascriptions of knowledge are claims to objective certainty, whether from the point of view of the subject or of the ascriber or, as usually, of both (who may be the same person). Like probability (indeed, it is certain that P if and only if there is no probability that Not-P), certainty is relative to the epistemic situation of some person or persons in question, including all the available evidence (although – unlike It was probable that P at tIt was certain that P at t, like I knew at t that P is incompatible with Not P). Note that the ‘evidence’ is not necessarily what is, or was, actually employed in support of a judgement, but what is or was available. Hence, new or newly considered evidence can be brought against a judgement of probability or certainty. All this explains what ‘special reasons’ or ‘defeaters’ are: ‘Special reasons’ effective against a claim to know that P are simply considerations in the light of which the probability that Not-P is seen to be greater than zero. The arguments of the sceptic simply fail this test. For the sceptic, anything goes.

All this is brought against so-called ‘fallibilism’ (or probabilism), ‘contextualism’, and, in Chapter 6, ‘externalism’, each of which is presented by its advocates as an appropriate – or the only effective – response to scepticism. The book ends with some consideration of the strengths and weaknesses of some recent anti-sceptical moves, in particular, argument on the question of how we know that our faculties are reliable, and the doctrine known as ‘disjunctivism’.

Maria Rosa Antognazza: Knowledge and the First Person

Drawing inspiration from Michael Ayers’s notion of “primary knowledge”, I argue that knowledge is essentially – that is, primarily, fundamentally, paradigmatically — first-person. To use a slogan, ‘each one of us has to do her own knowing’. Knowledge, in its primary and paradigmatic sense, is not a bundle of information that can be passed on as a parcel. Strictly speaking and at its most fundamental, knowledge cannot be transferred because each one of us has to see, grasp, perceive for herself (either literally in sense-perception or metaphorically in intellectual perception) the object of knowledge. In this regard, knowledge is similar to understanding – or, more precisely, understanding is a paradigmatic manifestation of knowledge in its primary sense.

I begin by summarizing some seminal philosophical accounts regarding the nature of mental phenomena and first-personal thought, focusing on key remarks by Brentano, Husserl, Hume, Berkeley, and Williamson. I conclude this section with Anscombe’s claim that the ‘I’ does not refer. In what follows, I try to unpack this insight by proposing an account of how we arrive, in the first place, at the distinction between ‘I’ and ‘not-I,’ ‘subject’ and ‘object’, within the broader framework of an investigation of what knowledge essentially, or most fundamentally, is. My proposal is that knowledge in its paradigmatic form is essentially first-personal in the sense that its object-directness requires a built-in awareness of the ‘I’ as the unifying perspective from which an aliud (a not-I, an object) is apprehended qua aliud. This is a first-order awareness which is crucially distinct from the second-order awareness which requires a reflective cognitive act – a distinction which I propose to cash out in terms of ‘first-person knowledge’ v. ‘self-knowledge’.

Thomas Crowther: Substances, Events and Processes

This paper focuses on the metaphysical background for the epistemology and philosophy of perception of Knowing and Seeing. A commitment of Michael’s metaphysics is that there are significant differences between the kind of unity possessed by substances, such things as human beings and lumps of rock, and the kind of unity possessed by events, such things as arm movements or rollings. Very roughly, there is a sense in which ‘anything goes’ with respect to the unity of events, in a way that it is not the case that ‘anything goes’ with respect to the unity of substances. 

This is a commitment I share. But in this paper I raise questions about this outlook that arise from reflection on the role that events, processes and activities play in understanding the identity of substances. I will go on to make some tentative suggestions about how we might resolve them.

Guy Longworth: Knowing, Knowing Perspicuously, and Knowing How One Knows

In Knowing and Seeing, Michael Ayers presents a view of what he calls primary knowledge according to which one who knows in that way both knows perspicuously and knows how they know. I’ll make use of some general considerations about seeing, knowing, and knowing how one knows in order to raise some (largely clarificatory) questions about this view. More specifically, I’ll consider (a) some putative limits on one’s capacity to know how one knows and (b) some putative functions of seeing, and remembering seeing, in sustaining what Ayers has (in previous work) called the authority of perceptual knowledge. The main question I’ll pursue will be whether perspicuity should be thought of either (i) as a condition of sensory experience, (ii) as a condition of sense-based cognition, or (iii) as an interface condition, involving interrelations between sensory experience and sense-based cognition.

Rory Madden: Perspicuity, Reflexivity, and Consciousness

I examine the striking claim made in Chapter 2 of Knowing and Seeing, that central cases of perceptual knowledge of the world involve (without any act of second-order reflection) knowledge that and how one knows.  I point out some parallel claims from the history of philosophy, ancient and 19th century, and suggest a way of elaborating more precisely the place of ‘consciousness’ in primary knowledge.

Charles Travis: In the Thick of It

This will be a discussion of three central points, and one less central one, in Michael Ayers’ book, Knowing and Seeing (2019). These are:

1. Through a rich and careful set of case studies, Ayers presents a prima facie case that sense can be made of the phenomenology of perception only on the assumption that the object of perception is always a mind-independent cohabited world. What there is to be seen, smelled, etc., is what there is to be seen, heard, or smelled, by one.

2. Knowledge granted by perception (in witnessing how things are) is primary, along perhaps, with some knowledge of conceptual truths. Other cases of knowledge are secondary. Primacy is in terms of ontological (and logical) priority: what it is for a secondary case to be knowledge must be understood with reference to the primary case. (Secondary cases need not be ‘second class’.)

3. Knowing-that is not a species of a wider genus, holding (taking to be) true. Unlike merely thinking-so, it is a direct relation to (taking-in of) a fact. (The fact is revealed, manifest, to one.) Knowing is not reducible to something else.

4. The less central point: Ayers rejects a disjunctivist account of knowledge, specifically that of John McDowell.

This essay aims to gain a better understanding of all these points by setting them in a Fregean context. For the first, it raises a question as to just what force phenomenology can have, and why this should be. For the second, it raises a doubt as to the importance of the distinction. For the third, the doubt is as to the aptness of genus-species talk for capturing the point. As to the fourth, in point of sociology, disjunctivism, and in particular McDowell’s, leaves many philosophers cold. McDowell’s presentation may bear non-zero responsibility for this. But more importantly, such philosophers have not looked carefully at what disjunctivism is. Again with reference to Frege I attempt to rectify the omission. There is a fundamentally important point here about method, and about two ways of conceiving a capacity.