Reason and Experience
DAA March 09

I. Mind as Tabula Rasa

The Specification:

- The strengths and weaknesses of the view that all ideas are derived from sense experience

- The strengths and weaknesses of the view that claims about what exists must ultimately be grounded in and justified by sense experience.

This is an analysis of the "empiricist" view: both Hume and Locke are empiricists as they argue that all knowledge depends on experience. Note that the first item asks us to evaluate empiricism as an explanation of the origin of ideas, and the second asks us to evaluate the claim that knowledge must be justified with reference to experience.

Locke on the origin of Ideas

AO1 Position and its implications:

The mind is a tabula rasa or "blank slate" at birth, empty of all ideas and knowledge; it is gradually filled through experience

AO1 Detail, Illustration:

An Essay Concerning Human Understanding II.1

Locke's definition of "idea" = "the object of thinking"

He gives examples: "such as are those expressed by the words whiteness, hardness, sweetness, thinking, motion, man, elephant, army, drunkenness, and others"

We might say concepts rather than ideas, the basic mental building blocks of propositions or declarative sentences

Ideas are acquired through two processes, sensation or reflection; each is a type of perception, the first of external objects, the second an inward perception of mental processes.

Sensation

The senses "convey" perceptions into the mind, or rather "they convey into the mind what produces there those perceptions". These are the ideas of "sensible qualities" e.g. colour, heat etc.

Reflection

Later on, once we have acquired some ideas through sensation, we are able to gain ideas from reflection; this is "the perception of the operations of our own mind within us." e.g. "perception, thinking, doubting, believing, reasoning, knowing, willing, and all the different actings of our own minds", which you can of course only begin to do once you have some ideas to think with/ about.

Simple and Complex Ideas

We can use ideas acquired from these two original kinds of perception to form other, more complex ideas:

"These, when we have taken a full survey of them, and their several modes, combinations, and relations, we shall find to contain all our whole stock of ideas; and that we have nothing in our minds which did not come in one of these two ways."

Objects in the world cause us to have ideas of them: "In the reception of simple ideas, the understanding is for the most part passive."

Locke's view is that there are no ideas at all that do not, ultimately, originate in sense experience.

He begins his "Essay" with a refutation of "innatism", the group of positions regarding our being able to acquire at least some knowledge independently of experience. His argument rests firstly on a rejection of "nativism" on the grounds that there are a whole lot of things that children don't yet know, but rests more importantly on a challenge to produce a simple idea that does not originate in either of the two kinds of perception; if we could do this, we could demonstrate that Locke is wrong.

Nativism (which Locke refutes) is the theory that we are born already knowing everything. It is not the only innatist or rationalist perspective: in fact, virtually no philosopher (except possibly Plato) has actually been a nativist. It's a pretty odd view. The other forms of innatism will easily escape Locke's refutation (see below).

AO2 Interpretation, Analysis, Application

Most of this can be treated under Hume (below) , but it is worth noting some of Hume's specific criticisms: he thinks that Locke is fuzzy in his application of the term "idea" and in his explanation of the process whereby we come to have ideas, and feels that his own definition, where an idea is "a copy of an impression", clears this up nicely.

Bear in mind that many of the terms, in particular perception and sensation, are used differently by Hume.

Hume on the origin of ideas

AO1 Position and its implications:

Ideas are (faint) copies of original sense impressions

AO1 Detail, Illustration:

Enquiry Concerning Human Understanding, Section 2

Because my original experience was forceful and vivid, it impressed upon me, like a stamp, a copy of itself.

Hume supports this with the claim that if we lack the ability to have the original sense impression, we can never form the corresponding idea. This explains why blind men can have no idea of colours and deaf men of sounds.

I can combine simple ideas to form complex ideas. Thus I can have ideas of things which do not actually exist, like golden mountains. I acquire this idea by combining my idea of gold, which comes from a real original sense impression of gold, and my idea of a mountain.

Note that although this is an idea of something which does not actually exist, and will therefore never be experienced, it is meaningful: it can be understood because it is composed of other ideas of things that we have experience.

Hume claims that the imagination is limited and is really no more than "the faculty of compounding, transposing, augmenting, or diminishing the materials afforded us by the senses and experience."

Both ideas and original sense impressions are called "perceptions" by Hume. The former is much less vivid than the latter, as it is only a faint copy. That means that if we wish to clarify an idea, we refer to an appropriate impression. In other words, when we want to understand what a dog is really like, we go and look at one. The sense experience itself is clear and cannot be mistaken.

We also have a means of determining whether our ideas are of things that actually exist: we go out into the world and look for a corresponding sense impression.

Hume's Fork

Hume goes on to make a broader claim about the justification of claims to knowledge. There are only two kinds of knowledge claim: relations of ideas and matters of fact. If we find a claim that does not rest either on a relation of ideas or an original sense impression (e.g. a theological claim...) we "Commit it to the flames: for it can contain nothing but sophistry and illusion." Relations of ideas are important, but they are restricted to mathematical and logical principles and do not actually add to our knowledge of the world.

A.J. Ayer and the "logical positivists" employ a version of Hume's fork. It is Ayer who adds the language of "analytic" and "synthetic" statements. His "verification principle" is a criterion for discerning whether synthetic statements, what Hume calls "matters of fact", are meaningful.

|Relations of Ideas |Matters of Fact |
|e.g. mathematics, logical rules |e.g. claims about the nature of the world |
|ascertained through deductive reasoning |ascertained through inductive reasoning |
|a priori - known independently of experience |a posteriori - known through/ after experience |
|analytic - true by definition |synthetic - true by correspondence to reality |
| |contingent - a truth that could have been otherwise |
|necessary - true in all possible worlds | |

You should prepare some examples of knowledge claims which you could use to demonstrate your understanding of the implications of Hume's Fork.

AO2 Interpretation, Analysis, Application

Circularity?

If Hume is right that all ideas depend ultimately on original sense impressions, we should not be able to produce an idea which cannot either be linked to an original impression or broken down into component ideas that can. Hume has fun showing how one common candidate, our idea of "God", can be broken down in this way: "The idea of God, as meaning an infinitely intelligent, wise, and good Being, arises from reflecting on the operations of our own mind, and augmenting, without limit, those qualities of goodness and wisdom."

Descartes' "trademark argument" for God's existence, on the other hand, rests on his claim that the idea of God as an infinite being must have its origin in some innate knowledge of God, because no empirical experience (given that such experiences are always of finite objects) could furnish us with the concept of infinity.

According to Hume, any idea which does not have a corresponding sense impression, or is not composed of other ideas that do, is either meaningless or confused. He argues that causation is one such concept.

This possibly weakens his position, as Hume echoes Locke's challenge that we could prove him wrong if we could only produce and idea that did not have a corresponding sense impression. But then he also says that the criterion for an idea being meaningless or confused is not being able to produce a corresponding sense impression - thus ruling out any candidate for proving him wrong!

So it goes with causation. Hume shows that we have no original impression that could correspond to the concept of causation. Rather than concluding, then, that there is one idea at least which does not come from experience, he claims that the idea of causation should be rejected as meaningless. In other words, when the imagination does attempt to furnish ideas independent of sense experience, it becomes confused. To sharpen up our language here: what is produced is not technically an idea at all - it is a senseless word. So Hume doesn't actually talk about "meaningless ideas" - rather he talks about not actually having the idea you thought you had.

The missing shade of blue

Hume actually considers an exception to his rule: his example of a man with all but one of the different shades of blue laid out before him; Hume accepts that he will be able to form an idea of the missing shade even if he has never had an impression of it. Hume's response is that this one exception is so "singular" that it does not show that his idea does not in fact hold in every other instance. That's actually a pretty weak response.

However, there are some other responses available. Perhaps the man actually forms the complex idea of the missing shade from real impressions he has had: the idea of the shades of blue he has seen and the more general idea of gradations of shade. We might also question whether he does in fact have an idea of the missing shade. What would this idea be? The colour actually being visible to "his mind's eye"? We will have more to say about this odd picture theory of ideas later on.

However, this discussion of colour is a good place to exploit a more serious objection to empiricism. Consider having a room full of green and yellow objects. According to the tabula rasa approach, I can from these objects alone form the ideas of "green" and "yellow". Think about that: I group the yellow objects together and apart from the green ones and learn from them the idea of yellow. Before I had the idea of yellow, I somehow applied an ordering principle which allowed me to mentally separate the yellow from the green objects. Surely that demonstrates that my idea of yellow comes prior to my experience of objects as yellow?

Scepticism about the external world

Hume accepts that there is no way of telling whether our sense impressions resemble the corresponding objects in the external world, or even (by extension) whether there is an external world at all. This implication possibly leads to an extreme form of scepticism (philosophical doubt) about the existence of an external world. This is disappointing for a theory of the formation of concepts which is attempting to pave the way for a justification of our claims to knowledge about precisely such an external world.

Note, however, that this is not a problem for all empiricist accounts, just for the ones that make use of "sense data". Sense data are the contents of my experience which are produced by objects in the external world (what Hume calls "impressions"). Impressions actually act as a barrier between my ideas and the external world that my ideas are "about". Other philosophers have cut out the concept of sense data and argued that in experience we encounter the world directly.

Communication

Hume's sense impressions are incorrigible and personal. I cannot ever "get them wrong", because they are just my own raw experiences, but by the same token I can never get yours, because they are yours. This leads to a problem when we are using words to communicate with each other and expect our meaning to be understood. According to both Locke and Hume, my word for "red" corresponds to my idea of red, which is in turn derived from my sense impression of redness, which is personal and cannot be shared with you. If I cannot know whether your sense impression resembles mine, I cannot ever know when I talk about red (or anything else) whether it invokes in you the same idea as it invokes in me, and thus whether you have actually understood me.

Before we get too panicked here, we should note that what this objection really shows is that Hume and Locke's account of what it is to be understood, or of the meaning of language, is beginning to look wrong.

Solipsism

If we combine the problem of scepticism about the external world with the problems of communication, we see that Hume and Locke's account possibly implies solipsism: the view that our mind is the only thing that exists, or at least that we only have access to our own mind and can never communicate with other minds. Hume and Locke's accounts have then been shown to lead to absurdity: however, that's not enough to reject empiricism - solipsism might just be the case. But common sense says otherwise, so we really need to look for an account of the acquisition of ideas that doesn't have these implications. Wittgenstein's linguistic turn (below) will probably do the job here.

The role of language

Hume and Locke both have kinds of "picture theory" of ideas, where words attach one for one to ideas which are themselves images or pictures of sense impressions. To learn the word is to learn the name for a particular idea, which attaches to a particular object.

Wittgenstein’s contribution here is to focus on the role of language in acquiring concepts. We shall have more to say about “language games” later on, but for now it is sufficient to note the later Wittgenstein’s contention that sense impressions are neither necessary nor sufficient for the acquisition of ideas. We say “the later Wittgenstein” as – earlier in his career – Wittgenstein originally had a picture theory of language.

The example of the “beetle in the box” is found in the Philosophical Investigations. The point is that we can happily understand each other’s talk of beetles even though we can never see into each other’s box: the point being that our private sense impressions can therefore play no part in our understanding of each other’s speech; it does not make much sense then to require them to have ideas.

“The thing in the box has no place in the language-game at all; not even as a something: for the box might even be empty. –No, one can ‘divide through’ by the thing in the box; it cancels out whatever it is.”

Wittgenstein argues against the “ostensive” account of learning language, i.e. that we are shown an object and hear the object named, and we come to associate the object with the name. He argues that sense impressions alone could not be sufficient for understanding the relation of the word to the object. Rather, the acquisition of language is more like being inducted into a public activity (or game) where we learn the rules about which things are appropriate to say in which situations (or “forms of life”).

So, to use the example from The Blue and Brown Books, if a speaker of a different language were to hold up a pencil and say “This is tove”, the sense impression could not guide us in our understanding of what “tove” meant: it could mean pencil, wood, round, one, hard or a whole lot else. It is in fact education into a shared and public activity that enables us to understand talk about pencils.

This approach leads the later Wittgenstein to reject Hume’s fork. There are in fact a variety of different language games that accompany a variety of forms of life, each with their own rules about how language can be used and when it is appropriate to say something is true. Meaning is embedded within the language game and cannot be separated from it and given a single definitive description. So Wittgenstein does not share Hume’s contempt for “metaphysical” language, as you will see to be significant in your later studies of the Philosophy of religion.

The implications for Hume and Locke’s empiricism are that we can understand or have an idea of something without having to have a mental picture of it, and words are not simply labels for things.

To have a concept (philosophers nowadays tend to say that rather than an idea) might be more like having a set of abilities, or an implicit awareness of the rules that govern when a particular word can be used.

If sense impressions are neither necessary nor sufficient for having an idea, then it is difficult to see what role they could play in clarifying obscure or faint ideas or in justifying claims to knowledge. Rather, what an idea is about can only be understood in the context of a shared way of life.

II. Innate Knowledge

- The strengths and weaknesses of the view that the mind contains innate knowledge regarding the way the world is: the doctrine of innate ideas and its philosophical significance.

Some (but not many) rationalists believe that all facts can be known independently of experience. This is not essential to being a rationalist - all that is necessary is that (unlike Hume and Locke) you believe that some knowledge about the world can be known independently of experience.

Plato

AO1 Position and its Implications

All knowledge is innate

AO1 Detail, Illustration

The most convincing argument for innatism is the argument from the acquisition of knowledge about colours that I have detailed above, but Plato gives an alternative in his dialogue the Meno.

In this dialogue Plato talks to a slave boy about geometry. The boy has never been taught geometry, but is able to produce a sophisticated theorem in response to Socrates only asking questions. Socrates says this could only be possible if the boy already had this knowledge, but had forgotten, and Socrates' questions triggered his recollection.

In the Phaedo Plato gives the example of our concept of equality. We know that the best we will ever observe in nature is two sticks that are almost equal in length, as nothing is ever perfectly equal. But to know almost equal presupposes that we know what "equal" is, and we could never obtain this idea from experience.

You should situate any discussion of Plato within his wider epistemology of the Forms and Platonic dualism:

1. Plato observes that the physical world is characterised by instability and impermanence; it is in a constant state of flux and change. In addition to this, we only ever observe particular objects e.g. "this cat", but we nevertheless claim to know what "a cat" is, and we are somehow able to recognise particular cats even though no two cats are alike.

2. For Plato, real knowledge must be perfect, eternal and unchanging; it must then be prior to our experience of an imperfect physical world.

3. Plato surmises that there is a perfect metaphysical realm of forms or pure ideas. It is through "participating" in (some say "resembling", but it is not clear how an object could "resemble" its form) these forms that objects in the physical world can be known for what they are. So in order to recognise any cat, we must already know what a cat is. We will not know that we possess the form of a cat until we have an experience of seeing a cat. This process of "recollection" of the forms is called anamnesis.

4. We possess perfect knowledge of the forms because the spiritual or mental aspect of the self existed in a disembodied state prior to birth in pure contemplation in the realm of the forms. After death, we return to this state before we are reincarnated. The process of embodiment causes us to forget that we have this knowledge, although it does not entirely erase the knowledge as experience can trigger it.

5. Objects can be said to be good in accordance with how well they resemble their particular form. Any experience, then, can help us to recollect the Form of the Good, which is the highest Form and in which all other Forms, in that they are good, participate.

6. While embodied in the physical world, we never arrive at a state of perfect recollection, but we can come closer to it the more time we spend in contemplation. The philosopher therefore has the best claim to real knowledge.

AO2 Interpretation, Analysis, Application

Some mathematicians are Platonists on the grounds that numbers are not something that we ever experience: we only ever see objects grouped in pairs, for example, we never see "two" but rather an instantiation of twoness. For Platonists, numbers are real entities that we can come to have knowledge of independently of experience, and it is in fact this a priori knowledge that enables us to count two objects. I suspect that Mr O'Boyle might be a Platonist about numbers; ask him.

The empiricist account of numbers, on the other hand, is that our knowledge of "two" is developed precisely from our experience of things grouped in pairs. The concept of "two" is an abstraction from several impressions of two things. For empiricists, to know two is not, in any case, to have any knowledge of the world. Mathematics is just a matter (as Hume would say) of relating ideas - it does not on its own yield matters of fact. I am pretty sure that Mr Smith favours this empiricist account.

There are plenty of ways of responding to Plato's rather odd claims, most of which problematise the concept of the Forms. Exactly what are there forms for? Is there a Form of cancer, in which case there could be a cancer which would be close to perfect and would therefore help me to recollect the Form of the Good. Is there a Form for a "Cat" or a Form for a "Siamese" and a "Manx". Is there a different Form for each different colour of a specific breed? If we restrict the number of Forms, it is hard to know how I can correctly distinguish a Manx from a Siamese. If we allow many particular Forms, it is hard to see how the realm of Forms is now distinct from the physical world of imperfect flux and particularity.

What exactly is the relation of "participation" in a Form? Forms cannot "look like" anything. If they did, they would take on particular aspects and could not stand for every individual instance of what they are a Form for. So a Cat's "catness" seems to have nothing to do with any of the particular things we associate with being a cat, as a cat without whiskers would still be recognisable as a cat...

However, rejecting the Forms still leaves us with that problem of how we correctly come to know cats as cats. For the empiricist, the concept of a cat is a complex idea, an abstraction formed from various particular sense impressions of cats. But then how do we know, without this a priori concept, that all of the things we grouped together to abstract the concept from were actually cats? It seems on the empiricist account that a cat could be whatever I decide it is. But we do seem to agree on what is a cat and what is not. This probably leads us towards Wittgenstein's linguistic turn again, where to correctly recognise a cat is to know the rules whereby in particular instances it is appropriate to speak of cats. This still leaves some deep mysteries unsolved which we won't develop now as even Wittgenstein gets quite anxious about not knowing the answer to them...

Leibniz

AO1 Position and its Implications

We have an innate capacity for forming knowledge that is triggered by experience

AO1 Detail, Illustration

Leibniz rejects the "blank slate" hypothesis and likens the mind to "veined marble", where the veins are the potential knowledge we can come to acquire but experience is the artist that hews that knowledge out of the marble. A piece of veined marble will lend itself to being carved in particular ways.

So we already have the potential to know everything we know about the world, but particular experiences trigger that knowledge. This can explain how even though knowledge is innate we can in fact know different things - different experiences have triggered different "veins" in the marble.

AO2 Interpretation, Analysis, Application

Contemporary biology, psychology and even linguistic theory does lend some support to Leibniz's view. The philosopher Peter Carruthers argues that humans have genetically determined capacities: the development of the eye is genetically encoded over different stages of our lives; similarly, perhaps certain ideas can only be developed after certain stages of development, and need not themselves come from experience. What is important is a combination of the child being at the right stage of development in combination with them being exposed to the appropriate stimuli.

Chomsky famously argues that we genetically possess a "deep grammar" which is a prerequisite for learning language. Without this "deep grammar", no amount of exposure to the speech of our parents alone would allow us to learn language.

Leibniz's view would solve the problem of the acquisition of colour concepts discussed in preceding sections. For Leibniz, we innately possess the potential to acquire the concept of yellow and this is triggered by our experience of yellow objects.

It is odd to think that we already have the idea of yellow but need an experience to activate it. That would suggest that our minds stored an almost limitless amount of dormant information. If, on the other hand, the mind just possesses certain capacities or categories for experience, we begin to move towards Kant's view, discussed in the third section.

- The view that some fundamental claims about what exists can be grounded in and justified by a priori intuition and/ or demonstration

Descartes

AO1 Position and its Implications

Certainty about the world must be based on the intuition alone, on the analogy of mathematics and geometry.

AO1 Detail, Illustration

Set out in the Meditations, argument spread over 6 chapters. Descartes wishes to examine all of his beliefs and find which he can hold for certain. Image of his beliefs as apples in a basket. Rather than test each one to see if it is rotten he intends to tip out the whole basket and start from scratch. He therefore resolves to impose the most severe test of doubt possible and see if any of his beliefs can withstand it. This will then be the foundation on which he builds his knowledge of the world.

FIRST MEDITATION

The three stage method of doubt: i) Descartes has sometimes been deceived by his senses i.e. he once saw a round tower from a distance and thought it was square. But there are methods by which we can distinguish false perceptions form veridical ones so ii) Descartes has sometimes dreamt and thought he was awake. Since there are "no reliable signs" by which - at the time - we can distinguish being asleep from being awake, Descartes must doubt now whether he actually knows anything about his situation, given that he cannot be sure he is not actually asleep. But even dreams are constituted of real perceptions we have had when we are awake, so even if Descartes must doubt the particular sense impressions he is having at the moment, there must be some things he knows i.e. that there are in general colours, shapes, objects, which he cannot doubt, so iii) the malevolent demon hypothesis. Descartes proposes to suppose that some omnipotent demon is deceiving him about everything he thinks he knows, even the truths of mathematics.

SECOND MEDITATION

Descartes now employs the image of being lost at sea without any firm ground. It is hard to see what beliefs at all could escape the doubt arising from the demon hypothesis. Descartes does, however, discover his "Archimedean point": his own existence. Elsewhere, Descartes claims "cogito ergo sum", but in the Meditations, he says: "this proposition I am, I exist, is necessarily true each time it is expressed by me, or conceived in my mind"

In other words, if he is thinking, even if he is doubting, then he cannot be deceived about at least one thing: that he exists, even if at this stage it is only his mind, or his essence as a non-extended "thinking thing" that he knows the existence of.

From this certain realisation, Descartes makes an observation about certainty: that he is certain whenever he has a "clear and distinct perception". Clear and distinct perceptions are incorrigible at the time you have them, as they are the very essence of certainty. But what about just after you've had one? Couldn't the evil demon make you think you'd had a clear and distinct perception in the past when in fact you hadn't? This would mean that Descartes' quest for certainty is stymied again.

In this section Descartes also sets up rationalism as the model for the acquisition of knowledge. He considers how he knows this "piece of wax". The wax has a certain shape, and smell, a certain response to touch, a certain appearance. Yet when it is melted, it can change every one of its sensible properties and still Descartes knows that it is the same wax. How does Descartes really "know" this wax? His explanation is that he does not know it through the senses but "grasps" it in his intellect alone. Its real properties are not therefore those that are linked to the senses but those that can be known through the intellect alone, i.e. the mathematical and geometrical property of extension, or taking up space. He further argues that intellect often trumps experience, such as when we see the sun it appears to be small and close but we actually understand through reason that it is an immense and distant sphere.

THIRD MEDITATION

Descartes presents his less well-known argument for God's existence, the trademark argument, which I touched on in section I above. There is some disagreement as to whether this is an a priori or an a posteriori argument, although I am pretty sure it is a posteriori and therefore illegitimate for Descartes at this stage. What is important is that Descartes thinks he is proving God's existence independently of experience.

Now that Descartes is certain that God exists, even it is just while he is thinking about it, he can validate the Clear and Distinct Perception. If God is a good God, he will not allow an evil demon to let him think he has had a Clear and Distinct Perception when he hasn't.

FOURTH MEDITATION

A slight digression where Descartes considers why, if God won't allow him to be deceived, he often makes mistakes. He gets God off the hook by arguing that God has by necessity given him a finite intellect, but that he occasionally sins by coming to conclusions about things when he has not yet had a Clear and Distinct perception.

FIFTH MEDITATION

As if he already knows the trademark argument is dodgy(!), Descartes offers his second, more well known, argument for God's existence, the ontological argument. He argues that it is a necessary part of God's being or essence that he exists, in the same way that it is a necessary part of a triangle that it has internal angles that add up to 180 degrees. In the same way that you can't have the idea of a triangle without three sides, you can't have the idea of God without him existing. You'll look at this argument in more detail next year. Suffice to say at this point that although this is a genuinely a priori argument, it doesn't work.

SIXTH MEDITATION

Descartes crams in the rest of his project: his arguments for mind/body distinction and his argument for the existence of the external world, as well as his answer to the first and second stages of his method of doubt. For our purposes, he also says a little more about how real knowledge is linked to the intellect alone: he compares his idea of a triangle to his idea of a chiliagon (a thousand-sided shape). He can picture or form an image of the triangle "in his mind's eye", but he cannot do so with the chiliagon. However, Descartes thinks that he nevertheless fully understands a chiliagon. This shows that the capacity of forming a mental picture is linked to the imagination, and ultimately the senses and thus the body, whereas understanding is rooted in the intellect alone. To know a chiliagon is merely to understand its geometrical properties.

AO2 Interpretation, Analysis, Application

FIRST MEDITATION

Descartes actually does not take his doubt as far as he could take it. If the demon could even deceive him about mathematics, it could even deceive him about the logical principles on which the rest of his arguments are based. But Descartes does not appear to doubt these principles, unless he is in fact attempting even to trump logic with the Clear and Distinct perception (which would lead to further problems, see below).

Wittgenstein's linguistic turn enables quite a sophisticated response to Descartes' demon hypothesis: the language in which Descartes frames his doubt is in fact publicly acquired via other physical beings in an external world. If Descartes claims to be doubting such beings, he must doubt the very foundations of the language he is using to frame that doubt. So it is not clear that such a global doubt can be meaningfully expressed in the language we are forced to use. For philosophers who use Wittgenstein's public language theory, Descartes' demon hypothesis is meaningless, which is quite handy, because none of Descartes' solutions to the problem are any good!

SECOND MEDITATION

Hume claims that the existence of the self is in fact very difficult to "prove"; what is this self of whose existence we are so certain? Hume says that he always falls back on particular empirical observations - this hand, this thought etc.

A.J. Ayer develops this response: the existence of the self has not been demonstrated a priori here. It is not a logical contradiction to say "I do not exist", as my existence is not a necessary truth. I may very easily not have existed. It is certainly an odd thing to say, as by its very utterance it demonstrates that it is false. But this is a pragmatic, not a logical contradiction. In other words, it can only be seen to be a contradiction because I actually exist. But this is presupposing my existence - the argument becomes question-begging.

It is very odd to say that the wax is grasped "by the intellect alone". How does the geometrical property of the wax communicated from the world to my intellect except via the senses? Elsewhere Descartes claims that this is down to a process somewhat akin to constant divine intervention. It is also strange to say that the real property of the external world is extension alone. It seems counter-intuitive.

THIRD MEDITATION

It would seem, if Descartes is working on the analogy of mathematics and geometry, that he need only to present deductive arguments in order to be certain about something. Why then does he need the clear and distinct perception, what is one, and how do I know I am having one? There are all sorts of problems with this, the main one being that I can only be sure about them once I know that God exists, but God's existence is established on the grounds of a Clear and Distinct perception. This huge gaff is known as the "Cartesian Circle".

FIFTH MEDITATION

We could spend quite a lot of time savaging this argument; suffice to say that what Descartes is trying to establish here is a synthetic a priori truth: God's existence is a matter of fact, but Descartes wants to know it through the intellect alone. Hume would argue that there cannot be synthetic a priori truths - reasoning alone can only yield relations of ideas, not matters of fact. So contemplation on the idea of God might show us that the idea of God contains necessary existence, but it cannot show us that God actually exists. To acquire knowledge of matters of fact, you need to go out into the world and have some experiences. You would need to search for a sense impression that corresponded to this idea of God.

- Is “certainty” confined to introspection and the tautological?

Spinoza

Spinoza also attempts to establish God's existence a priori through an ontological argument. Once we know that God exists and has created the world, we know that there are in fact no contingent truths. Leibniz argues that since God has created the best possible world, everything that happens happens necessarily and in accordance with his will. For Spinoza, since God creates the causal principles according to which all events happen, all events happen necessarily in accordance with his will.

The implications of this are that you could, if you had sufficient (infinite?) knowledge, deduce every event in the universe from your initial understanding of God and his goodness. This claim resonates with Hawking's claim that if we could understand the theory of everything, the simplest equation through which he would express all of the workings of the universe, we would know the mind of God...

For Spinoza, truths only appear contingent to use because we do not possess the perfect understanding of the mind and will of God to see them as necessary. If Spinoza's ontological argument doesn't work, this whole thing doesn't get off the ground, but the Hawking idea possibly points to a way of developing Spinoza's argument without God.

Most philosophers today accept that deductive reasoning is limited. To use Hume's distinction, we don't acquire any facts from playing around with relations of ideas - we need to go out into the world and have some sense impressions to do that.

So where does that leave the project of acquiring certain knowledge? Well, relations of ideas are certain, but that it is because they are tautologies or true by definition. Can we attain certainty through other means?

Well, there is a problem with the inductive process through which we acquire matters of fact or scientific knowledge. Whenever we generalise from particular instances to every instance, we risk error. It is a common misconception that inductive arguments are based on probability - they are not. However many white swans I have seen, that does not make it a jot more "probable" that the next swan I see will be white.

There is a further problem with induction which Hume makes clear. We cannot justify the use of induction deductively, so it would seem that any attempt to justify the process of induction must be based on inductive reasoning itself. This would seem circular. Hume's response is that we use induction out of force of habit but we cannot justify it. Quine argues that we have an "instinct" for induction and that we don't justify instincts, we just act on them. I would add that just because induction cannot yield certainly knowledge, that doesn't necessarily mean that we don't have good grounds for using it. Some philosophers of science have argued that we do not in fact acquire scientific knowledge through inductive grounds, but I have now run out of space for this section :o)

III. Conceptual Schemes

The Specification:

- The idea that experience is only intelligible as it is, because it presents sensation through a predetermined conceptual scheme or framework; and the philosophical implications of this view.

Kant's A Priori Categories

AO1 Position and its Implications

A synthesis of rationalism and empiricism: certain a priori categories are required for experience, but there is no a priori knowledge of the world as there is no knowledge of the world without experience. Our concepts provide a "conceptual scheme" which shapes our experience of the world

AO1 Detail, Illustration

Kant's realisation has been called the "Copernican Revolution" in epistemology. It's pretty flipping significant.

Remember how we questioned in section I whether we could know that other people had the same "sense impressions" as us. Hume and Locke's empiricism seems to depend on us having some kind of raw, incorrigible (can't be wrong) sense datum which we then copy into an idea. It is at the point of copying that things can get blurred or we can make mistakes.

The text book (Atherton et al, p.5) cites an anecdote from Gombrich to make an important point here. The German artists try to "transcribe what they saw with the utmost fidelity" but see that "their transcripts differed to a surprising extent." This perhaps demonstrated that there is perhaps no such thing as a raw, uninterpreted experience.

Lacewing says that to have an experience "of" or "about" something is to have an experience that has been interpreted or mediated. He argues that an unmediated or uninterpreted experience would be just a "buzz" about which we could say nothing. Kant's contribution to this debate is to argue that this would not be an experience at all. He distinguishes "noumenon" - the world "in itself" outside of the structuring processes of the human intellect (this world is inaccessible to human beings) - and "phenomenon", the world structured through experience or the experiences themselves (the two are, of course, indistinguishable).

So there are no experiences apart from experiences "of" or "about" things, and to have an experience of or about something requires having a category through which to interpret it. This process does not get off the ground unless there are at least some a priori categories which we possess innately, before having any experiences. These categories do not provide knowledge of the world, so Kant is not at this point an innatist - the categories are the a priori prerequisites of experience, which we must have before we can acquire any knowledge of the world. But Kant is not an empiricist, because although we cannot have knowledge of the world without experience the categories themselves are not abstracted from experience but exist in the mind prior to it.

Kant thus argues that we can have "synthetic a priori knowledge" - the categories are not known through experience but they are nevertheless not analytically true. In fact, we need some experiences before we can even know that we have these categories, but the categories must be innate.

The categories:

Categories of quantity: unity, plurality, totality

Categories of quality: reality, negation, limitation

Categories of relation: substance/ accident, causality/ dependence, community/ interaction

Categories of modality: possibility-impossibility, existence-non-existence, necessity-contingency

One way Kant argues for the necessity of categories is in his discussion of causality. Remember Hume argues that we have no sense impression which could correspond to our idea of causation? Well, Kant agrees, but whereas Hume says that this is therefore a confused idea which we should ditch, Kant argues that causation is one of the essential a priori categories which make our experiences possible.

Kant explains that sometimes the order in which our experiences occur is significant and sometimes it is not. For example, I might enter my house from the front door and have a series of perceptions, e.g. bathroom followed by (as I walk down the hall) lounge followed by kitchen. On another occasion I enter through the back door, and perceive kitchen followed by lounge followed by bathroom. The order of my perceptions does not matter here - my understanding of the house has not changed, and neither has the house.

However, on another occasion I stand by a river and see a ship first small and later big. Here the order of my perceptions does matter. If I had seen the ship as first big and then small, I would know that it is travelling away from me, whereas with the order of my perceptions being as it is I know that it is travelling towards me. According to Kant, only the category of causation, the concept that effects follow causes in a necessary temporal order, could allow me to make such a distinction. But I could not abstract this category from experience, as without this category I could never have had an experience of a ship travelling toward me or away from me.

AO2 Interpretation, Analysis, Application

Kant's argument for particular categories does seem pretty convincing. Possibly the categories are part of our genetic inheritance as human beings, which is why we are able to have experiences.

There is a pretty major problem, however, which is that contemporary scientific thought rejects some of Kant's categories. Well, okay, maybe he was just wrong about some of them. If we think that we could perhaps "fix" Kant's list and agree a concept of categories that actually do order human experience, then we pretty much agree with Kant that there are some a priori categories.

Contingent/ Linguistic Conceptual Schemes

AO1 Position and its Implications

However, have another look at the passage from the Gombrich. Is it possibly the case that the categories through which we experience the world might differ from culture to culture or person to person? So that different people or cultures might in fact have radically different kinds of experiences? That would be to acknowledge that the categories were contingent; that although they precede and prefigure my experience, they are not actually a priori and absolute but somehow tied to the particular cultural or linguistic situation in which I find myself. To understand that we need to return to some of the linguistic insights of the twentieth century.

AO1 Detail, Illustration

Sapir

"The 'real world' is to a large extent unconsciously built up on the language habits of the group..."

The implication of this claim, based on an anthropological study of cultures that speak different languages, is that different groups of language-users experience the world differently; some have gone as far as to say that not only is their experience relative to the conceptual scheme, but reality is in fact relative, but we'll come back to that later.

This claim rests on the principle that certain concepts are untranslatable across cultures. Many of the examples cited in support of this (like the famous one of the Inuit having so many different words for snow) are just plain apocryphal. If a term cannot be translated word for word, that does not mean it is untranslatable - it just might require more words in the other language to convey the same idea. But that does not mean the idea is untranslatable. Think about it: even if I need to write an essay to convey all of the senses of the word ennui, and still haven't exhausted them (something which, incidentally, I may not even be able to do for a word in my own language, but we don't normally require this to say that I understand the concept) does that mean I have never experienced the emotion that corresponds to what the Frenchman calls ennui? I simply don't believe that a Hopi Indian would not very quickly find ways to express whichever of our temporal concepts Whorf claims are unavailable to him, and not just because to learn the language would be to learn the concepts. Many philosophers don't see how his experience of time could be so radically different to ours. But I think that this idea might have some mileage when it comes to complex scientific "paradigms" or ethical language.

Whorf

"...the individual is free to describe nature with absolute impartiality but is constrained to certain modes of interpretation even while he thinks himself most free."

The implication of this is linguistic determinism: this is a concept which is prefigured in Nietzsche's "bewitchment of language". Since we think in language, to learn a language is to learn how to think about the world. You might illustrate this with reference to the "Newspeak" with which the totalitarian government controls the thoughts of the populace in Orwell's 1984. For Nietzsche, to think differently about the world would require creating a new language, but this would be an act of terrifying intellectual iconoclasm which is not available to the ordinary human being.

Rorty

Contingency, irony and solidarity

You can use here your material on Wittgenstein's "language games" from Section I, as Rorty relies on this work. For Rorty:

"Truth cannot be out there - cannot exist independently of the human mind - because sentences cannot so exist, or be out there. The world is out there but descriptions of the world are not. Only descriptions of the world can be true or false. The world on its own - unaided by the describing activities of human beings - cannot."

Rorty's is a "postmodern" perspective, but it must be noted that he is not as radical as it is sometimes supposed he is. Rorty does not claim that reality is relative to the speaker - only descriptions of the world are relative. The world in fact constrains the descriptions that are available to us, but it is the rules of the language game we are using which determine whether the descriptions are true or false.

I illustrate this by speakers of two different languages watching some people running around with a ball. The actions of the players (the "world") constrain the descriptions that are available to me, but it is the rules of the language game I am using which determine whether my descriptions are true. Consider that someone picks up the ball and kicks it. In the language-community of "rugby", my statement "That was a good kick" is true, whereas in the language-community of "football", my claim "That was a foul" is true.

However, the implication of this is that there are no criteria for determining truth outside of the specific language games - there is no right answer to whether that was a good kick or a foul. What could possibly mediate between the different language games apart from another language game? The language we use, and the rules for which claims are true and which false, is therefore "contingent", it is rooted in a particular cultural or historical context and does not carry absolute or objective force. For some people this is major problem as it means we would have no grounds for condemning an alternative morality, e.g. the morality of the Romans where it was acceptable to watch gladiatorial combats, rape our enemies and possess slaves: the best we could do would be to say that this form of life is out of fashion.

However, Rorty is not a linguistic determinist. He accepts that language games do evolve and we can in fact invent entire new ways of thinking. Humans, after all, shape the languages they speak. For Rorty, this is what occurs when there is a "paradigm shift" in science or a major religious revelation changes the way people look at the world. For Rorty, new scientific or moral language games are not "advances", discoveries that somehow move us closer to the way the world is than the knowledge we had before, they are alternative languages which have been created by poetically inspired individuals and caught on.

Rorty sometimes seems in danger of contradicting himself when he suggests that languages catch on because they are more "practical" than the alternatives. This would seem that there is some kind of objective meta-language by which we could assess the relative usefulness of individual language games for describing the world or performing particular actions.

However, Rorty in fact argues that the language game should not be separated out from the form of life, so that the scientist who conceives a new way of describing the world is in fact thinking of new ways of interacting with the world which could not be described in the language they used before. If people want to interact with the world in this new way, they will begin to use the language which accompanies the new form of life.

Rorty thus claims that the old model of a philosopher who sought to describe the world objectively should be replaced with a new model of philosopher as poet or literary critic, someone who is bravely trying to conceive new ways of being, or at least is commenting on knowledge as a critic interprets an artistic work, as a contingent human product.