20th World Congress of Philosophy Logo

Philosophy of Mind

Blindsight and the Role of the
Phenomenal Qualities of Visual Perceptions

Ralph Schumacher
Institut für Philosophie

bluered.gif (1041 bytes)

ABSTRACT: The aim of this paper is to defend a broad concept of visual perception, according to which it is a sufficient condition for visual perception that subjects receive visual information in a way which enables them to give reliably correct answers about the objects presented to them. According to this view, blindsight, non-epistemic seeing, and conscious visual experience count as proper types of visual perception. This leads to two consequences concerning the role of the phenomenal qualities of visual experiences. First, phenomenal qualities are not necessary in order to see something, because in the case of blindsight, subjects can see objects without experiences phenomenal qualities. Second, they cannot be intentional properties, since they are not essential properties of visual experiences, and because the content of visual experiences cannot be constituted by contingent properties.

bluered.gif (1041 bytes)


Blindsight is often understood as supporting certain claims concerning the function and the status of the phenomenal qualities of visual perceptions. In this talk I am going to present a short argument to show that blindsight could not be understood as evidence for these claims. The reason is that blindsight cannot be adequately described as a special case of seeing. Consequently, it is not possible to draw inferences from it concerning the role of the phenomenal qualities for seeing.

Visual perceptions are supposed to have two sorts of content. First, they have intentional content which relates them as representations to the external world. The properties that constitute the intentional content are called representational or intentional qualities. Second, visual perceptions have phenomenal content which characterises the subjective experience of them. The phenomenal content is constituted by introspectively accessible sensational or phenomenal qualities, so-called visual qualia.

Concerning the role of the phenomenal qualities for seeing particularly the following two questions are of importance:

(1) The first question is dealing with the causal or functional role of phenomenal qualities: Under the assumption that seeing is based on cortical information-processing, the question arises, whether the phenomenal qualities of visual perceptions have a function with regard to this processing, in the sense that the intentional content of visual perceptions depends not only on their intentional, but also on their phenomenal qualities. Is it true, as among other authors Frank Jackson and Steven Pinker claim, that phenomenal qualities are only epiphenomena, not having any function for information-processing? (1)

(2) The second question concerns the status of phenomenal qualities: Are the phenomenal qualities of visual perceptions non-intentional qualities, or do they belong to a certain type of intentional qualities? In other words, are phenomenal and intentional contents essentially different, or are the so-called phenomenal contents of visual perceptions really part of their intentional contents?

In order to come to a decision concerning these two questions, several authors, among them Peter Carruthers, Daniel Dennett, Colin McGinn, Nicholas Humphrey and Robert VanGulick, refer to an empirical phenomenon called "blindsight". (2) Since blindsighted persons receive informations without conscious perception, blindsight is often understood as supporting the claim that the phenomenal qualities of visual perceptions do not have a function for the information-processing which gives rise to their intentional contents, and, consequently, that they are not intentional qualities. In other words, since blindsight is understood as evidence for the possibility of visual perceptions of having intentional content, but lacking phenomenal content, blindsight is interpreted to show that visual qualia are epiphenomenal with regard to the information-processing of visual perception.

The Phenomenon of Blindsight (3)

Blindsight is a defect which some persons suffer from as a result of a partial damage of their primary visual cortex. It is typical for this partial cortical blindness that there is a blind area within the visual field of the persons, who have this sort of brain damage. Blindsight is remarkable because of the contrast between the lack of declarative knowledge of the blindsighted persons about the stimuli presented within the blind area, and the high rate of correct answers to questions concerning those stimuli. On the one hand, blindsighted persons claim to see nothing within their blind areas: they say that they do not to have any visual experiences of objects presented within the blind field. In accordance with this, they are not able to react spontanously to optical stimuli presented in the blind area, and they also cannot decide on their own, whether something was presented to them in the blind field, and how the presented objects look like. But on the other hand, with a frequency significantly above chance level they are able to give correct answers concerning opticial stimuli present within the blind area, when they are asked to decide between given alternatives. However, this ability is restricted to a narrow range of optical stimuli. For example, blindsighted persons can give mostly correct answers about the presence and absence of points of light, and they are also able to distinguish simple geometrical forms and patterns. With regard to optical stimuli of this kind the rate of success of blindsighted persons is between 80 to 90 % of the rate of success of persons with normal sight. (4)

It is particularly important that blindsighted persons describe themselves as not seeing anything within their blind fields, and as not having any knowledge about the objects presented there. Hence, from their perspective they are only guessing, when they are forced by the insistence of the experimentator to make decisions between given alternatives. What is amazing about blindsight, is the fact that there is a significant contrast between the declarative and the procedural knowledge of the blindsighted persons.

Blindsight as a Special Case of Visual Perception

We now come to the question of whether blindsight can be understood as a special case of visual perception. The reason for this examination is that only under the condition that blindsight can be regarded as a certain type of seeing, it is possible to draw conclusions concerning the role of the phenomenal qualities for seeing from the characteristics of blindsight.

Under the precondition that there is a sufficiently illuminated object x within the visual field of a physiologically intact person s, the following three points are candidates for necessary conditions of the application of the term "seeing":

(1) contrafactual condition: The occurrence of a visual perception v is contrafactually dependent on the presence of an object x within the visual field of a person s, in such a way that s would have had another visual perception v´ with a different content than v, if instead of x the different object y would have been presented to s.

(2) condition of consciousness: The visual perception respectively its phenomenal qualities must be experienced consciously.

(3) epistemic condition: If a person sees something, then she acquires beliefs about the presented objects.

ad (1): The contrafactual condition was developed by David Lewis in order to exclude veridical hallucinations from seeing. This condition is fulfilled in the case of blindsight, because the answers of blindsighted persons vary in dependence from the objects which are presented within the blind field. (5)

ad (2): If it is true that blindsight is characterized by the absence of phenomenal consciousness, then it does not meet the condition of consciousness. Nonetheless, this condition is not without problems, because it excludes all types of unconscious visual perception. But psychological research has shown in a number of different cases that the contents of unconscious visual perceptions of persons with normal sight actually do have a systematical influence on their behaviour. Therefore, as a general condition for seeing the condition of consciousness is too narrow and has to be replaced by a less restrictive condition, which allows both conscious and unconscious visual perceptions. Consequently, we may regard blindsight as a case of unconscious visual perception.

ad (3): Verbs of perception are mainly used in two different ways: (1) In the case of the non-propositional or nominal use, the grammatical object of the verb consists in a noun, as for example in the sentence "I see Jane". Generally, this use can be characterized with sentences of the following form: "s perceives x" ("s" and "x" are placeholders for persons and for perceived objects). (2) In the case of the propositional use, the grammatical object of the verb is a proposition, as for example in the sentence "I see that Jane is blonde". This use can generally be described as follows: "s perceives that x is F" ("s", "x" and "F" are placeholders for persons, perceived objects and properties).

In the distinction between non-epistemic and epistemic seeing Fred Dretske and also G. J. Warnock and Frank Jackson refer to these two different uses of perceptual verbs. (6) According to Dretske, non-epistemic seeing is a basic form of visual perception which does not imply the acquisition of beliefs. Non-epistemic seeing is described by sentences of the general form "s sees x". It is characteristic of non-epistemic seeing that from "s sees x" it does not follow that "s sees that x is F". Hence, it is possible to describe the content of non-epistemic seeing independently from the beliefs of the perceiver. That means, "s sees x" is an extensional context, where it is possible to substitute coreferential terms for "x" salva veritate. In contrast to this, epistemic seeing is characterized by the acquisition of beliefs. Generally, it can be characterized with sentences of the form "s sees that x is F". The content of epistemic seeing must be described with reference to the beliefs of the perceiver. Thus, "s sees that x is F" is an intensional context, to which the above mentioned principle of substitution cannot be applied.

Blindsighted persons describe themselves as not having any knowledge about the objects presented within their blind fields. Hence, blindsight does not fulfill the third condition, if we rely on an internalist conception of belief, since according to this conception we can ascribe beliefs only under the condition that, for example, a person has second-order beliefs about her first-order beliefs, (7) or that she is able to articulate her beliefs verbally, or at least that she does not deny to have those beliefs we are ascribing to her.

But if instead, following Charles S. Peirce or Gilbert Ryle, we assume the externalist conception of belief that it is sufficient for the ascription of beliefs, if persons have certain dispositions of behaviour, then blindsight may meet the epistemic condition. (8) Although, it is still necessary to define the range of behavioural dispositions in a way that allows to include the dispositions of persons to react with guessing when responding to forced-choice-tests. In other words, we also have to take into account those dispositions of behaviour which persons do not show spontanously, but only after being forced to make decisions between given alternatives. On the one hand, this externalist conception of belief is tailor-made for blindsight. But on the other hand, it is an extremely wide concept of belief, and, thus, not convincing. Therefore, the answer to the question of whether blindsight meets the epistemic condition, must be negative.

To sum up, from what has been said we can draw the conclusion that blindsight may at best be regarded as a type of unconscious and non-epistemic seeing. But in this case the distinction between the intentional and the phenomenal content cannot be applied to blindsight. This distinction is reserved for epistemic seeing, because non-epistemic seeing has neither intentional nor phenomenal content. We don´t even have to go so far as David Armstrong, who claims that only those perceptions have intentional content, which give rise to beliefs. (9) We only have to follow Dretske´s view that non-epistemic seeing carries no information. (10) Therefore, since there are good reasons against ascribing intentional content to blindsight as a case of non-epistemic seeing, we come to the conclusion that blindsight cannot give us any evidence about the function of the phenomenal qualities with regard to the intentional content of visual percep tions.

In addition, it is also doubtful whether blindsight actually meets the conditions for non-epistemic seeing. According to Dretske, if a person sees a certain object non-epistemically, then this object is visually differentiated from its immediate environment by the perceiver. (11) In other words, the object looks some way to the perceiver. Since blindsighters report that there is nothing which looks some way to them, it is not without reason to conlude that blindsight even fails to meet the conditions for non-epistemic seeing.

So we may conclude that blindsight is neither epistemic nor non-epistemic seeing. Therefore, I think that what we cannot learn from it, is something about the role of phenomenal qualities for seeing.

bluered.gif (1041 bytes)


(1) F. Jackson (1982): Epiphenomenal Qualia. In: Philosophical Quarterly, Vol. 32, 127 - 136; S. Pinker (1985): Visual Cognition: An Introduction. In: S. Pinker (ed.): Visual Cognition. Cambridge/Mass., 1 - 65.

(2) P. Carruthers (1989): Brute Experience. In: Journal of Philosophy, Vol. 86, 258 - 269; D. Dennett (1991): Consciousness Explained. Boston et al; N. Humphrey (1993): A History of the mind. London; C. McGinn (1991): The Problem of Consciousness. Oxford; R. VanGulick (1989): What difference does consciousness make ? In: Philosophical Topics, Vol. 17, 211 - 230.

(3) L. Weiskrantz (1997): Consciousness Lost and Found. Oxford et al.

(4) A.J. Marcel (1983): Conscious and Unconscious Perception. In: Cognitive Psychology, Vol. 15, 197 - 300

(5) D. Lewis (1986): Veridical Hallucination and Prosthetic Vision. In: D. Lewis: Philosophical Papers. New York et al., Vol. II, 273 - 290

(6) F. Dretske (1969): Seeing and Knowing. London, 4 - 77; F. Jackson (1977): Perception. A Representative Theory. Cambridge/Mass., 154 ff.; G.J. Warnock (1956): Seeing. In: Proceedings of the Aristotelian Society, Vol. 55, 201 - 218

(7) D. Armstrong (1968): A Materialist Theory of the Mind. London

(8) C. S. Peirce (1986): How to make our Ideas clear. In: Writings of Charles S. Peirce. C.J.W. Kloesel (ed.), Bloomington, Vol. III, 257 - 276; G. Ryle (1949): The Concept of Mind. London, Chapter 5

(9) D. Armstrong (1968): A Materialist Theory of the Mind. London, 209 ff.

(10) F. Dretske (1969): Seeing and Knowing. London, 77

(11) Dretske (1969), 20 ff. (See footnote 11)

bluered.gif (1041 bytes)


Back to the Top

20th World Congress of Philosophy Logo

Paideia logo design by Janet L. Olson.
All Rights Reserved


Back to the WCP Homepage