Epistemology

Page 1

Epistemology is the study of knowledge. There are several epistemologies, according to which branch of philosophy someone is studying:

• Virtue epistemology

• Social epistemology

• Moral epistemology

You can see from the titles that epistemology is related to many different topics that people study – for example, someone studying sociology might study social epistemology, and someone studying politics might study virtue epistemology; so it is not unique to people studying philosophy.

When we learn about and interact with the world we use our senses – for example, we touch a flower, we smell it, we see its colour, we can feel if it is warm or cold outside in the sunshine, and so on. This is known as “sense-knowledge”.

However, there are other aspects to the world that are outside of our senses. And sometimes, such as when we see an optical illusion, our senses are not telling us the truth, so it’s clear that there are ways in which our senses can be deceived.

Page 2

And so, although we know what we know about the world through what we experience about it, sometimes what we have experienced and continue to experience, or have thought and continue to think, just might not be what we thought it was. So, is this because what we see and know of the world constitutes what the world really is, or is it just the way that the world appears to be? And, if that is the case, are what we see as truths, for example, the colour “orange”, appearing differently to each person or not?

These are all thoughts and debates that have held philosophers’ attention for thousands of years. When a baby is born, it does not appear to have any knowledge other than reflex activity which ensures it gets fed, it gets its nappy changed, and it is kept nice and warm and cuddled from time to time. Its whole being is about satisfaction of basic need. The baby’s brain is what philosophers call a “tabula rasa” – in effect, a blank sheet upon which, as it goes through life, knowledge and ideas, opinions and prejudices and so on will be inscribed.

Page 3

The baby will learn lots of things before he or she reaches old age and dies. It will learn how to do things; this is called “ability knowledge”, and includes things such as riding a bicycle, being good at maths, playing a musical instrument, learning a trade and so on. Once we have the ability to do something, we tend to do it without thinking too much about how to do it. I have the ability to touch-type, which is great when it comes to writing this course, because I do not have to look at the keyboard and I can type quite fast. But, if I had not used a keyboard before, I would not have that ability knowledge and it would take me a long time to even type out a paragraph.

Another type of knowledge is known as “acquaintance knowledge”. This is a form of knowledge that you learn for yourself, without help from an intermediary, such as another person, a news item, a book, a film etc., telling you that knowledge. Learning this course is not acquaintance knowledge, because the intermediaries are the course notes and your tutor. Acquaintance knowledge would be something like knowing what the beach looked like on your last holiday because you experienced it for yourself. You know what colour the sand is, what the sea felt like when you swam in it, and so on. You did not get the knowledge from a brochure, but learned it directly from your own experience of actually being there. Acquaintance knowledge is knowledge of an object; you know the physical object but, using the beach analogy as an example, you do not know the truths of it. The truths might be knowing which and how much of each chemical dissolved in the sea is present, the precise number of grains of sand on the beach, and so on.

Page 4

Acquaintance with something does not consist of forming any judgment, or even thinking about it, nor does it mean having any concept or representation of it. You cannot be acquainted with something that does not exist. However, you can conceive of a concept that does not yet exist – which is how inventions, paintings etc., are brought about.

Propositional knowledge and the tripartite view

However, to know that something is knowledge, it has to be a ‘justified true belief’. So, in order to answer this we have to know four things:

1. What are the conditions of that knowledge?

2. What are the sources of that knowledge?

3. What is the structure of the knowledge?

4. What are the limits of the knowledge?

Page 5

Knowledge can also be defined as “justified belief”. This means that we need to know:

1. How we understand “justification”

2. What makes the justified knowledge, justified

3. Is the justification internal or external to our own mind

Propositional knowledge follows the “s knows that p” process, where:

S = the subject who has the knowledge

p = the known proposition.

Traditionally, p is a justified true belief in which:

1. We cannot know a false proposition, therefore knowledge needs truth

2. If s does not believe p, then it cannot be something that s knows. Therefore, knowledge needs belief

3. It might be that s believing p being true is down to luck, rather because they have a good reason to believe it. Therefore, knowledge needs justification.

This is known as the tripartite view, because there are three things that are essential if p is to be a justified true belief.

Justification: this is a word that means that knowledge cannot be down to luck, it has to be certain. There are two strands of knowledge, the first is traditional knowledge (TK), and the second is non-traditional knowledge (NTK). Both of these agree that we need to ask how justification can fulfil its role.

Page 6

Traditional knowledge says that

S’s belief of the truth of p is not due to luck, but has to be rational or reasonable from s’s point of view. If this is the case, we can take p to be true

Non-traditional knowledge says that

S’s belief needs a large degree of objective probability to be the truth. If this is the case, then p is not true because of luck. And, this can only happen if a belief stems from reliable cognitive processes or faculties (this is known as RELIABILISM)

So, we have

1. S is justified in believing that p

2. p is true and

3. S believes that p

The above conditions are the ‘three parts’ of the definition of knowledge, such that if S has a justified and true belief of p, then S knows p. This is why they are termed ‘necessary’ and ‘sufficient’. If S knows p, then it is necessary that S can fulfil all three conditions, and if S can fulfil these three conditions, then they are sufficient to know that p.

However, we know that 1) if p is false 2) if S does not believe in p, and 3) if S cannot believe that p is justified, then S does not know p.

We now know that the tripartite view of knowledge equates justified true belief with knowledge, and that knowledge is not knowledge unless it is justified true belief, and that justified true belief is all you need for knowledge.

Page 7

There are several things we need to ask ourselves about the tripartite definition of knowledge, because it seems to have several weaknesses. Let’s see what they are and take the following propositions:

1. Justification is not a necessary condition of knowledge

2. Truth is not a necessary condition of knowledge

3. Belief is not a necessary condition of knowledge

We can look at each one individually:

Justification

Think of the premise that someone holds an unjustified true belief. It would be possible to believe p, where p was true, but this is on unjustified grounds. Let’s consider the following situation:

1. S could think that T is a murderer because T is wearing a purple coat.

2. T just happens to be a murderer – so the proposition (p), that ‘T is a murderer’, is certainly a true belief

3. However, S is unjustified in believing p, since ‘wearing a purple coat’ is not evidence for this particular p.

4. Therefore, p is not justified in this case.

I’m sure you can think of further examples where true beliefs can be held on unjustified grounds, so perhaps this means that justification should not be one of the conditions of knowledge? But this in itself is problematic – surely knowledge does need some evidence or justification for itself, or else how can we determine what does and does not count as knowledge?

Page 8

There are several things we need to ask ourselves about the tripartite definition of knowledge, because it seems to have several weaknesses. Let’s see what they are and take the following propositions:

1. Justification is not a necessary condition of knowledge

2. Truth is not a necessary condition of knowledge

3. Belief is not a necessary condition of knowledge

We can look at each one individually:

Justification

Think of the premise that someone holds an unjustified true belief. It would be possible to believe p, where p was true, but this is on unjustified grounds. Let’s consider the following situation:

1. S could think that T is a murderer because T is wearing a purple coat.

2. T just happens to be a murderer – so the proposition (p), that ‘T is a murderer’, is certainly a true belief

3. However, S is unjustified in believing p, since ‘wearing a purple coat’ is not evidence for this particular p.

4. Therefore, p is not justified in this case.

I’m sure you can think of further examples where true beliefs can be held on unjustified grounds, so perhaps this means that justification should not be one of the conditions of knowledge? But this in itself is problematic – surely knowledge does need some evidence or justification for itself, or else how can we determine what does and does not count as knowledge?

Page 9

Truth

Can we think of knowledge as justified belief, rather than justified true belief?

It is possible for S to believe in p, which is not true, though S may both think and feel certain that p is the case.

We all do this. Look at the next page for an example:

How many times have you thought that today was Thursday, and in fact it was Wednesday? Or you were really sure that you had a dental appointment next Friday, but in fact it is not until the following Friday?

So we can say that we had a belief in ‘p’ — that ‘today is Thursday [when it is Wednesday] BUT we know that this false belief cannot also count as knowledge - because knowledge must be true. In which case, should you say that you only believed that today was Thursday, rather than that you knew today was Thursday? But, you believe that today is Thursday, in your mind you are sure you know that today is Thursday.

So should we only say that we believe (rather than know) any p until we can determine that p is definitely true?

Page 10

Belief

What if S knows p, where p is true and justified, and yet S does not believe that p is true? Here’s another example.

Let’s suppose that S is taking part in a quiz, and is asked a question about Number One Hits of the 1970s, and this is a subject that S has revised really well, so S is sure s/he will know the answers. Although S correctly answers a question about the Christmas Number One in 1975, S does not actually think that this is the right answer because s/he is feeling nervous and hasn’t settled into the competition yet.

We now know that S does not believe p — the proposition that “Bohemian Rhapsody by Queen” was the 1972 Christmas Number One in the UK —even though p is both true and justified, since S has based p on evidence-based revision of the subject.

In which case, we can see that S knows p, but does not believe p. So what about the conditions for the tripartite theory now? S cannot justify his/her answer, but the answer just happens to be justified. In which case, S does not know p – it might LOOK like S knows p, but this is just a happy coincidence.

Maybe we can look more closely at the nature of belief and say that S does in fact believe p, just not on a conscious or immediate level; it was a subconscious belief, but they still believed it even if they didn’t think they knew it. Therefore, S believes p, and we can therefore say p does still equate to knowledge.

Page 11

Issues which arise from the Anthology Focus

Gettier’s core challenge against justified true belief equating to knowledge is about his separation of ‘justification’ from ‘truth’.

This means that justified true belief can no longer be sufficient for knowledge, and if this is the case, it is no longer the same as knowledge. These cases are known as ‘Gettier cases’, in which knowledge cannot be justified true belief.

It means that the justification condition needs to be modified, removed or replaced in order to maintain the tripartite definition of knowledge. There are four main responses to Gettier’s challenge:

1. Strengthen the justification condition

2. Add a ‘no false lemmas’ condition (J + T + B + N)

3. Replace ‘justified’ with ‘reliably formed’ (R + T + B) = reliabilism

4. Replace ‘justified’ with an account of epistemic virtue (V + T + B)

Page 12

No false lemmas

This is the easiest of the four cases to understand. Go back to the link and read Gettier’s example of Smith and Jones.

Smith has a false belief that Jones will get the job, and he doesn’t know that the man who will get the job has 10 coins in his pocket. In which case, we need to add an additional condition to exclude the possibility of false beliefs so that we can ensure the following conditions:

1. The proposition p is true

2. S believes that p

3. S’s belief that p is justified

4. S did not infer that p from anything false

Page 13

In addition to the original three conditions, we have added a fourth. This fourth condition is known as the “no false lemmas” conditions. A lemma is something that you introduce into an argument in order to help you make your argument.

In this case, Smith based the proposition that Jones will get the job on the information given to him by the president of the company; and he also used this to decide that the man with 10 coins in his pocket will get the job.

Now we can see that “Jones will get the job” is a lemma, since it helps Smith to reach the conclusion that the man with 10 coins in his pocket will get the job – however, we also know that it is a false lemma. So it does not fit with the condition 4 that we added to the tripartite definition of knowledge.

This suits the Smith and Jones example that Gettier mentioned; however, we can see that it still doesn’t address the fundamental issue that truth and justification can be separated. We can only use this additional case where the Gettier case is one in which an inference was made, as Smith did.

But there are Gettier cases where there is no inference made, and so condition 4 is not broken, BUT justification and belief have still been separated.

So the tripartite definition of knowledge still remains problematic.

Page 14

Strengthen the Justification Condition: Infallibilism

What is justification? In the tripartite definition of knowledge we have no criteria for determining whether a belief is justified or not.

Smith holds two beliefs:

1. That Jones will get the job

2. That the man who gets the job will have 10 coins in his pocket.

But are these two beliefs actually justified? Should Smith believe the president of the company when she says that Jones will get the job? Smith might not be justified at all, or maybe he is not sufficiently justified in his belief. On the face of it, it might seem that the information given by the president of the company is very good evidence on which to base your own belief, but it is by no means certain.

Page 15

But are these two beliefs actually justified? Should Smith believe the president of the company when she says that Jones will get the job? Smith might not be justified at all, or maybe he is not sufficiently justified in his belief. On the face of it, it might seem that the information given by the president of the company is very good evidence on which to base your own belief, but it is by no means certain.

Infallibilism states that knowledge must be certain. It’s only infallibility that can provide the basis for justification, and can make true belief become knowledge. It is hard to be certain about something, so one conclusion to remember from Gettier cases is that we do not often have the sufficient justification for our beliefs to equate them to knowledge.

Now we need to think about the question, “what counts as infallibility?” or “how can we be certain that we are certain?” For this we need to look at Cartesian philosophy. Rene Descartes (1596—1650) was a French philosopher who is given the status of the “Father of Western Philosophy”. Much of what philosophers believe today are responses to what Descartes argued, and the term Cartesian philosophy applies to his philosophy. In 1619, he shut himself in a room with an oven to escape the cold, and had three visions which revealed a new philosophy to him which was unlike the philosophy of the past. He saw that all truths were linked to each other, and this gave rise to his famous saying that we have probably all heard of:

Page 16

“…I think, therefore I am…” (…cogito ergo sum…)

Descartes believes that reason is the means by which we reach certainty. If we do not always use logic and reason, we will begin to make mistakes. We need to remove ourselves from what our senses tell us because the senses can lie to us (think of optical illusions, for example).

1. One cannot know what is false (though one can believe what is false)

2. So if S knows p, then S must be sure and certain about p

3. In which case, for justification to make knowledge certain, justification must also be an indicator of truth

4. If S is justified in believing p, it is impossible for S to be mistaken

5. If there is any possibility of S being mistaken, then S is not justified in believing p

6. In which case, infallibilism [complete certainty] is truth

Page 17

Infallibilism helps us to exclude Gettier cases: as we know, in these cases there is no justified true belief if we use the infallibilism argument that knowledge equates to certainty. So, looking at the problem, we can believe that infallibilism supports the tripartite definition of knowledge.

Is there a problem with infallibilism itself though? In reality, we cannot often be totally certain based on the evidence we have in front of us. One of the consequences of infallibilism is that we can have a lot of beliefs which we could be fairly confident of being true – but we could know very little indeed, since there is very little we can be entirely and completely certain of. In which case, infallibilism does not help us to build up a quantity of certain knowledge, but instead leads us towards scepticism: we must doubt everything, since certainty is very hard to come by.

This is what Descartes did. He found that he could work out what was real by removing anything that he could doubt. This gave rise to a strand of philosophy called “Cartesian doubt”. His method included

1. accepting only information you know to be true

2. breaking down these truths into smaller units

3. solving the simple problems first

4. making complete lists of further problems

Page 18

He said

“…Several years have now elapsed since I first became aware that I had accepted, even from my youth, many false opinions for true, and that consequently what I afterward based on such principles was highly doubtful; and from that time I was convinced of the necessity of undertaking once in my life to rid myself of all the opinions I had adopted, and of commencing anew the work of building from the foundation...”

René Descartes, Meditation I, 1641

Replace ‘justified’ with ‘reliably formed’ (R + T + B) = reliabilism

Does knowledge actually need to have any justification at all; can it exist without it? Maybe it doesn’t - this is the concept of reliabilism, which states that you can claim to know something (p) if:

Page 19

1. p is true

2. You believe p

3. The belief is brought about by a reliable act of cognition

“A reliable act of cognition” is when you remember or you perceive something. These acts bring about true beliefs which can be considered knowledge even though they give rise to false beliefs. It might be that someone is mistaken in their beliefs, or they believe a lie which someone has told them believing. This is something that happens if, for example, there are a number of witnesses to a traumatising event.

Take, for instance, a car accident. Someone might believe that “the car hit the pedestrian”, and someone else might believe that “the car smashed into the pedestrian”. Both of these accounts contain truths, but the way that they have been expressed gives you two completely different beliefs of the same event.

Page 20

Reliabilism is useful in that it allows for non-human forms of knowing: when an animal interacts with its environment it knows what it is doing, for example. This means that we can assume that animals have a reliable cognitive process, so they must have true beliefs that count as knowledge.

But this form of reliabilism doesn’t help us with some of the Gettier cases: here’s a famous one about Henry, who drives through a place known as Barn County. Henry doesn’t know that nearly all of the barns in this county are actually fake – although they look real to Henry from his perspective, they have only a false façade at the front so they look real.

Each time Henry sees a barn he believes it is a real one, because that’s what barns look like to him. The twist is that there is actually one barn in the whole of Barn County which is real. When Henry sees this one, he thinks that he has seen another real barn: and he is right, but it’s only an accidental truth.

Page 21

Reliabilism does not help us because every time Henry sees a barn, he is using a very reliable cognitive process (his perception). When Henry sees the one true barn in Barn County, reliabilism would have to hold that Henry does in fact know that he is seeing the one true barn, but he has no idea that most of the barns he has seen are fake.

In other words, reliabilism does not help us when somebody holds a true belief under circumstances where that belief is only true by accident.

In which case, let us return to reliabilism and make it more nuanced, such that you know p:

1. If p is true

2. You believe p

3. This belief is brought about by a reliable act of cognition

4. You are able to determine what are ‘relevant possibilities’ in the actual situation

Henry is unable to determine that the barns are actually fake where they seem real, so he falls foul of condition 4 under this version of reliabilism.

Page 22

Replace ‘justified’ with an account of epistemic virtue (V + T + B)

Virtue epistemology is an extension of reliabilism; the term ‘virtue’, as it used here, relates to an intellectual ability which helps us to determine the truth.

There are two broad categories of virtue epistemology, which each have a different focus:

1. Reliabilist virtue epistemology looks at the virtues of cognitive abilities, such as memory; perception or rationality.

2. Responsibilist virtue epistemology looks at intellectual qualities, such as dedication to the truth, being unprejudiced and having intellectual honesty. It is more centred on the qualities of being a virtuous (not used in a moral sense) knower, rather than on the definition of knowledge.

Page 23

Reliabilist virtue epistemology holds that you know p:

1. If p is true

2. You believe p

3. This true belief is brought about as a result of intellectual virtues

This skills-based definition of belief gives us a definition of knowledge as something which arises from the practise of intellectual virtue, which in turn gives rise to appropriate beliefs. If we look again at Henry in Barn County, we find that Henry uses his intellectual virtues to form a belief that he has seen a real barn.

His belief is therefore suitable and appropriate, even then in the case of the fake barns, this belief cannot be knowledge because it falls foul of condition 1 – it is not actually true.

The reliabilist response is as follows: Henry doesn’t actually have the ability to determine which barns are real and which are false, because in the environment of Barn County, it is very difficult/impossible to tell which barns are real and which are fake from Henry’s perspective.

Page 24

Because Henry cannot know which is false and which is real, the true belief he holds when he sees the one real barn cannot be based on his ability to discriminatebecause of the context of Barn County. So, his true belief is not suitable and appropriate because it was not formed through intellectual ability.

Henry happens to be lucky when he holds his true belief about the one real barn; this also cannot be as a result of his abilities. In which case, the reliabilist position helps us to get out of this particular Gettier case. Henry did not exercise intellectual virtues in Barn County, and so his p ‘this is a barn’ when he see the one real barn, is not knowledge.

Earlier on we saw how it might seem natural to conclude that all our ideas come from sensory experience. The conclusion we reached is, effectively, the position of the empiricists, one of the most influential movements in modern philosophy.

Page 25

The nature of definition and how propositional knowledge may be analysed or defined

There is a problem in defining knowledge as valuable, which is known as the value problem. Socrates, an ancient Greek philosopher, raised the question as to why knowledge should be seen as of more value than true belief. We might feel that knowledge has the greater value, but if we look more closely the problem becomes more involved. Suppose you wanted to get to York. You might have one of two things in your mind

1. A true belief of what is the route to York

2. Knowledge of the route to York

If both will get you to York, is the knowledge of how to get there more important than the belief about how to get to York? The end result is the same.

The issue of why knowledge is considered more important than true belief is known as the Meno problem.

Page 26

The term “Meno problem” comes from Plato, who was also an ancient Greek philosopher. He argued that knowledge possesses a property which is tied to truth – this is similar to an ancient Greek myth about the statues of Daedalus, which were so life like they were tied to the ground to stop them from running away. For Plato, true belief was like the statues not being tied down and able to run away and get lost. He stated that knowledge, unlike true belief, gives us a certainty and confidence which is cannot be lost, and therefore knowledge is superior to true belief.

We can return to our example of how to get to York. If you believe that you can get there, your belief could “run away” with you because you might get confused – suppose the right road to York appears to be pointing in the wrong direction. Your belief might then lead you astray, because you would not have the knowledge that you must follow the road that appears to be going the wrong way. If you knew that, initially, the road appeared to be going in the wrong direction, you would be confident in knowing that it was OK at this point to appear to be going the wrong direction, because eventually you would come to the turn in the road.

But, as always in philosophy, someone could present an argument that supported the true belief scenario and refuted the knowledge scenario. This would be known as taking a revisionary response to the problem. If we accept the premise that knowledge is of more value than true belief, we are taking a non-revisionary approach, because we have not revised the argument to turn it around.

Page 27

The Secondary Value Problem

What (if anything) makes knowledge more valuable than that which falls short of knowledge? Duncan Pritchard is a professor of philosophy at Edinburgh University, and he wrote a journal article in 2007 entitled “Recent Work on Epistemic Value”. He questioned why we desire knowledge rather than any epistemic understanding that falls short of knowledge – this could include true belief.

We can suppose that knowledge = true belief + an additional component, to cope with the Gettier style cases. Also, what if we suppose that justification adds value to true belief? If knowledge includes justification, then it is of more value than true belief – but justified true belief, according to our equation can be equal to knowledge.

However, we cannot always assume that knowledge is of greater value than true belief. Suppose that you truly believe p, when p is the possibility of winning the lottery. And suppose you have knowledge of p, when p is something that is going to kill you. In this case, I am sure we would all agree that true belief is preferable to knowledge.

Page 28

Linda Trinkaus Zagzebski

Zagzebski (born 1946) is an American philosopher who works in the field of epistemology. In her book, Epistemic Authority: A Theory of Trust, Authority, and Autonomy in Belief (2012), she defends a strong sense of epistemic authority, including authority in moral and religious beliefs, and argues that belief in authority is a requirement of intellectual autonomy.

She argues that views that define knowledge, in terms of true belief, arise out of belief forming processes. Her argument is that the reliability by which we produce something doesn’t necessarily give it more value, and therefore we cannot assume that the reliability of the process itself will add value to the true belief. She uses the “cup of coffee” analogy:

Page 29

Criticism

Berit Brogaard (1970 - ) is a Danish/ American philosopher specializing in the area of philosophy of language amongst other topics concerning neuroscience. She is Professor of Philosophy at the University of Miami. She states that a valuable cause can confer value on the effect, in certain cases. She claims that virtue epistemologists, such as Zagzebski, adhere to what she has called a “Moorean” conception of value, which is where if two things have the same intrinsic properties they are equally valuable. So, according to this, if true belief and knowledge do have the same intrinsic properties, then it follows that they must have the same value.

Therefore, it is vital that we understand knowledge to have distinct intrinsic properties that are different from true belief, if the value problem is to be solved.

If we say that there are only intrinsic and instrumental values, then objects that have the same properties have the same value, both for their intrinsic and for their instrumental properties, unless there is something external to change one of the values. But this can create a problem, because there are some things (for example, the ring that my mother bequeathed me when she died) that have more value because they are related to something else we valued (as I valued my mother). The same ring, if I saw it in a shop and it had no connection for me, would not mean anything like as much to me as the one that my mother gave to me. We see this often, for example, when something that belonged to a famous now-dead person is auctioned and fetches many thousands of pounds, when to all intents and purposes, instrumentally, it is nothing like that valuable.

Page 30

http://osclinks.com/4792

If those letters had been from one ordinary, non-royal person to another ordinary person they would certainly not have sold for £15,000. They are non-instrumentally worthless, but intrinsically they are very valuable.

This means that we could value one true belief more than another one, because of its extrinsic properties. Maybe it came from a source which is itself independently valuable; we may believe a piece of information from one source more than we believe it if it came from another source. It might have come from a cognitive trait with value of its own, rather than believing something for its own sake – for example, if someone told you that Jesus was God and expected you to believe it, just because they told you so, it would be of less value than someone telling you that Jesus was God because they had concrete proof, and showed it to you.

Page 31

A key question in epistemology is the importance of what we see, hear, smell, taste and touch to what we know, and how we experience the world. This is what we call “sense data” – and we will be looking at this theme within this topic. We will look at the senses in more detail, ask ourselves exactly what it is our senses are telling us, and how these senses relate to what is going on around us in the outside world.

By the end of this section you should understand:

Direct realism

1. The immediate objects of perception are mind-independent objects and their properties

2. Issues arising from:

• The argument from illusion

• The argument from perceptual variation

• The argument from hallucination

• The time-lag argument

3. The responses to these issues

Indirect realism

4. The immediate objects of perception are mind-dependent objects (sense data) that are caused by and represent mind-independent objects

• John Locke’s primary/secondary quality distinction

5. Issues including:

• The argument that it leads to scepticism about the existence of mind-independent objects

6. Responses including:

• Locke’s argument from the involuntary nature of our experience

• The argument from the coherence of various kinds of experience, as developed by Locke and Catharine Trotter Cockburn (attrib)

• Bertrand Russell’s response that the external world is the “best hypothesis”

7. The argument from George Berkeley that we cannot know the nature of mind-independent objects because mind-dependent ideas cannot be like mind-independent objects.

Page 32

Direct realism

Most people will never have stopped to think about perception, and certainly not in the way that philosophers think about it. The theory of direct realism is sometimes known as naïve, or common-sense realism.

There is the view that what we experience every day by way of our senses IS the external world and what we see, feel, hear, etc directly is what is really going on and what is actually there around us. This is a strand of philosophy known as empiricism, developed in Britain in the seventeenth and eighteenth centuries by philosophers such as John Locke, David Hume and George Berkeley. It is a philosophical tradition that holds that all significant facts about the world can be derived from our own experience.

Objects around us have size, shape, colour, mass and so on, and these things we know to form part of the object itself. We see the object and experience it directly. We see green grass, blue seas, white clouds, grey skies, a brown dog, and many, many other things too. We just assume that this is all there is to it all. This is what is known as naïve realism. It is just an assumption of what is there without having to think about it. The sky is really blue, yet if you go into the science of why the sky appears to be blue, there is far more to it than literally meets the eye. This is fine… or is it?

Let’s think about this in more detail. Most people don’t think about it, and they assume that they have a common-sense view of what is all around them. It is fine for everyday life, but for philosophy it is unsatisfactory, and there are plenty of questions to be asked about it. We need to get to the truth of it all, because that is what we do as philosophers.

Philosophy causes us to question even the most basic of our beliefs to test them. If they are right, they will come out stronger than they were before. If they are not right, then we have the chance to abandon them. However, for direct realism the theory begins to fall apart when we scrutinise it closely.

Page 33

The argument from illusion

If you decided that you were a direct realist, you would trust your senses all the time. Your senses are designed to tell you about the world in which you live, so why not put your absolute faith in them? Descartes battled with this problem for quite a long time, and he showed that with the ‘argument from illusion’ we find that our senses are good at deceiving us in some situations. They tell us things about the world that are not right, and show us things that are not actually how the world really is. Here is a fun activity for you to do to test your sense of sight.

This first illusion is called the Hermann Grid Illusion. Look at the picture below and count all of the black dots that you see.

Page 34

I expect that you realise that there are no black dots in reality. Your brain is making them up.

Let’s try another one. This one is called the Kanizsa Triangle. How many triangles are there?

Page 35

And yes, there are no triangles, just three V-shapes and three Pac-man like shapes.

Here’s the final one.

Stare at the light bulb picture for 25 seconds and then look away at a blank, preferably white, surface. What do you see?

Page 36

This time you should see a glowing light bulb.

That’s enough of the fun part; back to work!

We have seen that even simple things can fool our visual senses, and earlier on you watched the video of the rubber hand experiment. So, if these senses are easy to fool, then it follows that other senses are also probably easy to fool. Plus, of course, our senses are linked. For example, in order to balance we use our eyes and the vestibular system within our ears; so if we disturb one of these senses we can become unbalanced – if you have had an ear infection you may well be aware of the dizziness and nausea that comes with it.

For example, if we look at railway tracks extending into the distance, it will appear that the two edges meet at a point on the horizon: however, we know for a fact that these edges run parallel to one another, in order for the trains to run along them, and therefore it is impossible that they meet. Artists use this illusion to show depth in paintings. Another example, offered by Descartes, is of a straight stick placed in water: if we look at the stick from the side, through the water, it will appear bent due to refraction of the light, as you may remember from your school science lessons. Of course, the stick is not really bent.

If we were using direct realism we might say that the railway tracks really do meet in the distance, or that the stick magically bends when we place it into the water, because our senses tell us that is what happens. However, we know from experience that these things are not really true. So, in this case we have no choice but to reject the common-sense view.

For the argument from illusion, there is also an ontological problem. Ontology is the branch of metaphysics dealing with the nature of being. If our senses are deceived, any attempt to deal with the world outside of our perception is rather futile – how can we rationalise the external world using senses that are easily fooled?

Page 37

The argument from illusion is a tool used by sceptics about the knowledge we have of the external world, and is designed on purpose to demonstrate that our senses cannot always be trusted. The sceptics argue from three main steps

1. What is it of which I am aware in my sense experience? We have already explored some of this in the optical illusion games. BUT in each case there is something that does have the sensory experiences that I am wrong about – it might be that the experience is of something small, or it was blurry etc. So, says the sceptic, there ARE sense data that are distinct from the self and from the world. There must be sense data, because otherwise there would be no sensory features at all.

2. It is best to give a single account of phenomena that are relatively similar. This means that misconceptions are dealt with in terms of sense data (step 1) so we should apply this to all experiences of perception, even if there is an error. These perceptions are known in philosophy as veridical perceptions (truth-telling perceptions)

3. Sense data mediates between the perceiver and the world. However, whatever they are metaphysically, they are intermediaries. And as intermediaries they can be true or false.

Page 38

Read Descartes, Meditations on First Philosophy, Meditation 1.

How does Descartes use doubt to question the evidence of his senses?

You can find the document at: http://osclinks.com/4794

If you have read the text and answered the question, you will probably have said something like Descartes telling us that we should reject all empirical evidence because, as he says:

“…it is prudent never to trust those who have deceived us, if only once.”

He also presents the analogy about trusting our senses being similar to trusting a friend who once deceived us; it’s the “once bitten, twice shy” argument, and of course it is sensible to not trust someone who has already lied to us once.

However, under most circumstances our senses seem to be fairly trustworthy – they have got us through life so far, and none of us are dead yet. I hope I am still alive when you are reading this!

Page 39

The argument from hallucination or dreaming

We’ve already had some fun with a few optical illusions, but now Descartes takes the matter further by saying that we might be fooled by all our senses at once. To do this, he uses the argument from hallucination (also known as the argument from dreaming).

The argument from dreaming is a bit of a puzzle. We may think that the argument of dreaming supports the idea that our sense data could be all that there is.

When we wake from a dream, sometimes we are surprised to find out that what seemed so real was just a dream. Think about your dreams at night. Some are vivid, some not so. Some are nice and you really enjoy them, and some are terrifying and really scare you. We dream of all these things and never once imagine that we are really tucked up safe in bed. We dream the most ridiculous of scenarios but to us, at the time that we dream them, they seem logical and real.

So, although we might be doing all sorts of different things inside our minds, our bodies are unaware of being asleep in bed. Your brain even makes sure that, unless you are a sleepwalker, your body is paralysed during sleep so that you do not harm yourself. This is a perfect example of every single one of our senses being fooled all at once. When we are asleep, our dreams are our reality, but when we are awake our conscious state is our reality, and our senses tell us all sorts of things that simply would not happen in the real world. We do not know that we are dreaming when we are asleep, and we do not think we are dreaming when we are awake. This relates to the philosophical explanation of the realist claim.

The sceptic might argue that this is a case of mistaking the non-veridical for the veridical. How can you ever be sure that you are not dreaming now, because when you were dreaming it was real to you? Maybe the sceptic is right, maybe nobody does ever know anything from the evidence of their senses. The sense data may always be the product of dreaming, even when we think we are awake.

Some philosophers argue that you can differentiate perception from dreaming, partly due to the greater coherence of perception. How, though, do we actually know that dreams have less coherence? They seem perfectly coherent to us when we dream. Maybe, then, perception is just a coherent dream.

Sceptics argue that whatever is the case cannot reliably tell us that something IS the case, because there is a difference between appearance and reality. If we want truth, we need to know what reality is. However, the only tool available to us is perception, and therefore we are liable to make errors.

We can support the sceptical claim with two arguments: ‘disagreement’ and ‘the brain in a vat’.

Page 40

Disagreement

In Protagoras, Plato notes that there is a wide range of perceptions. For example, you might be feeling cold in a room and the person next to you might think that the room is very warm. My husband I and I often have this disagreement! It’s common for people to argue whether something is blue or green, too. So how can we say that something is warm or cold, blue or green? We then get into the realms of subjectivism, and start to lose empiricism.

Disagreements come about because we believe that our perceptions are giving us knowledge about something. We can measure something, and then we have added objectivity. For example, we could measure the true temperature of the room, or use a colourimeter to determine the true colour.

The brain in a vat

Suppose we are not real human beings, just a brain in a vat connected to a supercomputer somewhere in the universe. The computer programmes my brain to think what it wants me to think, and when it wants me to think it. It gives me an illusion that I am real, but I am not. But I cannot know that, because everything, including my body, and the fact that I am typing this, and the sounds of the keys and the sunlight shining on me, are all false. Even memories are created by the computer, the past is programmed in, but it doesn’t exist at all. Even the logical judgements and mathematics are programmed in. I am not a free agent to judge anything, which means I cannot know anything.

Page 41

Let us consider hallucinations, these can come about as a result of physical disease processes such as dementia; sometimes people with dementia realise their hallucination for what it is, but as dementia progresses, the hallucinations become increasingly real until they cannot be separated from the individual’s personal reality. People who are isolated for long periods, often used as a form of torture, seem to develop hallucinations, and may hold conversations with imaginary people. But, in the main, most people never do experience a hallucination. Note also that hallucinations can be of various types, people see things, hear things, feel things and so on. People hallucinate when they suffer psychosis, or take certain drugs. They can also be generated by stimulation of the temporal lobe, by magnetism or electricity.

Hallucinations

1. For hallucinations we perceive a property, F

2. Because we perceive F, then something has F

3. We do not perceive a physical object

4. Therefore, what we perceive is mental; this is sense data.

Page 42

If we assume that people can have a hallucination when they are perfectly lucid, we cannot tell if the event occurred because it is a subjective experience on the part of the person who hallucinated. Therefore, we cannot hold that the argument from hallucination is true – also, how do we know that we are not hallucinating at the present time? Maybe we are, but it is being corrected for by our other senses.

Descartes put forward more evidence for his argument from hallucination. He said:

“… for is there anything more intimate or internal than pain? And yet I have sometimes heard people say, who have lost arms or legs, that they still sometimes felt pain in the limb which had been amputated; and this caused me to think that I too could not be quite certain that any one of my limbs was really affected, although I felt pain with it...”

(Descartes, Meditations on First Philosophy, mediation 6)

For Descartes, his doubt arises from the experiences of other people who have had limbs amputated. But have the other people recognised their pain hallucination for what it really is? They know that their limb was amputated, they know what it was like to have the limb, and they base their hallucination on what they previously knew about their bodies. So they know that the pain is hallucinatory and can tell it from reality.

If you could not distinguish a hallucination from reality, such as some dementia sufferers and people who have psychotic illness cannot, then it cannot be recognised as a hallucination like the people with amputations can recognise their hallucinations. Hallucinations are not just visual, as we know. But we need something to contrast the hallucination with to know that it is a hallucination; however, if the hallucination is not real, what can we use to contrast it with?

Page 43

Hallucinations and perceptions are not the same. Both can be concerned with memory but the former is a memory of something that doesn’t exist except in the mind, and the latter is a memory that we form of something that exists, that is tangible to more than one of our senses. However, the memory of tangible things and events can also be fooled – have you ever revisited a place that you remember from being a child? Quite often you will be struck by how small it is, or how it isn’t quite laid out how you remember.

Hallucinations show that subjectivity can be indistinguishable from veridical perceptions

1. So, in both hallucination and veridical perception we see sense-data.

2. Therefore we see sense-data and not physical objects immediately

3. Therefore direct realism is false.

So, this is the point of view as presented by Descartes. There are further scientific arguments that we can look at which should rule out the idea of direct realism completely. There are two that we will examine more closely, and these are

1. The argument from perceptual variation

2. The time-lag argument

Page 44

The argument from perceptual variation

The argument of perceptual variation takes the premise that the world that we see in physics textbooks is very different from the world that we see around us. For example, if you saw a sphere from above, you might think it was a flat circle. In other words, we perceive physical objects using our direct experience.

Russell said that a shiny brown table looks different colours depending on where it is you are standing. Maybe you are standing somewhere where the light is reflecting off the varnish; the colour will look different if you then come back and look at the table on a dull day, when the sun is not shining. So is the ‘brownness’ the property of the table, or does it share it with other colours that you might have seen on different occasions? It might look different if you were looking at it via a microscope – then the smooth surface would probably be very rough with little bumps that you had not seen before. It is clear that the perception has changed for my brain, but the table remains the same. Therefore, what we perceive is not the same as what exists in the mind, because we have seen differences to what we expected of the table. We know the table is rectangular at one angle, but at another angle we might have seen it as a distorted diamond shape; this would be known as a “relational property”, because the table looks more diamond shaped than rectangular, rather than looking as it was intended to be, a rectangle.

So, we have not perceived the object directly. In other words, we have perceived what our sense-data have told us.

Let’s take another example – a red tomato. We know how it tastes, and we know its texture. Some of us like tomatoes, some of us do not. But in the description of physics the tomato is made up of atoms, which have no taste (we think, since nobody has ever tasted an individual atom), no smell and no colour. Not only that, but an atom mostly consists of empty space (to be precise, 99.9999999999996% of a hydrogen atom is empty space), so how can the tomato exhibit its properties? How can a tomato even exist, if it is virtually all empty space?

If we now examine the individual atom we find protons, neutrons, electrons and the empty space. The electron is not even considered to be a fundamental particle – in 2012 researchers in Switzerland separated an Electron into two smaller quasi-particles – a “Spinon” which governs the spin, and an “Orbiton” which controls the orbit; there is also probably a particle called a “Holon” that carries the electron’s charge. The whole atom consists of positively, negatively and neutrally charged particles and empty space. Not quite like the way we experience a tomato! So is the tomato real, or just a construct of charges and empty space and of course some atomic bonds?

This violates the central premise of common-sense realism. It seems that the world isn’t anything like we perceive it to be. Because, if it were true, then we could perceive these atoms and bonds with the naked eye.

So, do we reject modern physics or do we reject common-sense realism?

We can use this geographically too. I could say that I am in Manchester, and therefore I am “north of” London. But it is only in relation to some other place. I could be “south of” Edinburgh, but I am still in the same place.

Refraction is another way that our eyes deceive us. Putting a pencil in a glass of water, then observing it from the side, tells us that there is a kink in the pencil. But is there? In this case the illusion is “subjectively indistinguishable”, and we have a veridical perception of it.

1. We perceive that the pencil is crooked, so this is the property, F. When we perceive F, the object, O, has this property

2. In an illusion, O does not have the property, F, and is not crooked

3. Therefore, F is sense datum.

4. So, in illusions we have sense data rather than physical object data, and we can say that in illusions we see sense data and veridical perception

5. Therefore, the conclusion has to be that direct realism is false.

Page 45

Considerations

1. Relational properties:

• Although the pencil appears crooked, it is not. The water has given it the crookedness, therefore it just looks crooked

• And sometimes we perceive how things look, sometimes we experience that they have properties that do not look as they did when we perceived them

• So we perceive a) the object and b) the properties.

The second argument that we look at is called the time-lag argument.

The time-lag argument

This argument springs from the fact that there is a time lapse between an event occurring and our personal perception of that event. If we look at a television programme, there is a small time-lag between the time that the broadcast is sent out and when we watch it on the screen.

The light from the sun is also subject to a lag – an 8 minute delay to be precise. And the light that we see from many stars in the sky at night come from stars long gone; for some stars the light left the star before the age of the dinosaurs, but we are only seeing it now. This means that the common sense view really cannot be correct; we do not see objects as they are. Let’s modify direct realism to indirect realism (indirect realism is also known as representative realism).

Complete this activity before moving on.

Page 46

The disjunctive theory of perception

This theory says that if something just looks a certain way (e.g. the varying colours of Russell’s table) then one of two different things is happening

1. I directly perceive a mind-independent object, F

2. Or (e.g. hallucination) it appears to me that F is there, but there is nothing that is F there at all.

And furthermore:

1. Although hallucinations and veridical perception appear to be the same, and some of our theories have appeared to prove that they are, there are two different types of mental state going on, so they can seem the same but there is no direct proof that they are the same, because we have to consider the two different mental states.

2. Hallucinations do not tell us about perception because they are imagination, not perception.

3. Therefore, the argument from hallucination is not an objection to direct realism.

Page 47

How to read a philosophy article - reminder

Ideally, scan through the work first, then make a closer reading and note down any points that you feel are relevant. Philosophy texts can seem daunting at first, but if you take time you can pick up the meaning; sometimes this might mean reading a sentence several times and meditating upon it, and sometimes it means doing a search of words that you have not come across yet. Putting things into your own words, and as I mentioned earlier on, pretending that you are explaining something to a colleague also taking this course, are both good ways of ensuring that you have understood something.

You should now be able to:

1. Distinguish between naïve and philosophical direct realism

2. Know about the problems posed for direct realism by the arguments from perception and hallucination

Indirect realism

Looking at the previous work on direct realism, we have now challenged the premise that we can believe what our senses are telling us. However, we have not thus far abandoned realism altogether.

Page 48

It is clear that although on the surface of things realism makes sense, philosophically it is somewhat inadequate. If we abandon it, though, we then move down the path towards scepticism, in which we eventually find ourselves doubting the existence of any real world at all. This is because scepticism is when we believe that our usual justifications for determining truth are inadequate, and if this is the case we can then assume that we do not have knowledge. And so scepticism is a form of challenge rather than a theory.

John Locke

John Locke FRS (1632 –1704) was an English philosopher and physician, widely regarded as one of the most influential of Enlightenment thinkers, and commonly known as the "Father of Liberalism". He tried to clarify the relationship between material objects’ qualities and our sensory experiences of them.

John Locke argued that ‘qualities’ are the powers that a physical object has to produce a sensation in the mind. For example, whiteness, blackness, coldness, and warmth are all ‘qualities’.

Locke gave objects two distinctive types of quality, which he called:

1. Primary qualities: the qualities that are impossible to separate from the object and that remain constant no matter who perceives it. For example, we all observe refraction when a stick is placed at an angle in some water. Such properties include size, solidity, and state of motion or rest, and our experience of them closely resembles the physical object that causes the experience

2. Secondary qualities: these are the qualities we perceive that tend to change depending on the object’s circumstances – for example, under a red light a white ball will appear red.

Locke stated that there was:

“…nothing in the objects themselves but power to produce various sensations in us by their primary qualities…”

(1690, Book 2, Chapter 8)

Locke pointed out that a violet doesn’t contain a property of “being blue” and a fire doesn’t contain a quality of “causing pain”. If you cut yourself with a knife the knife doesn’t resemble the pain it caused you when you cut yourself. These are what he describes as secondary qualities.

He clarifies his points by using something called the “argument from variable perception”. With this argument, he shows that the secondary qualities are not part of an object and are relative to whoever perceives them.

Let’s take a look at a possible scenario to explain this point. You have two hands, and you put one in a bowl of hot water and the other in a bowl of iced water. A third bowl contains lukewarm water and, after a minute or so you take both hands out of the respective bowls and put them both into the lukewarm water at the same time. What happens? The lukewarm water will feel hot to one hand and cold to the other. So, from this experiment we can see that the secondary quality of the lukewarm water is not a constant, because it feels different not only to different people, but in this case it actually feels different to the same person.

Page 49

The first bite of the apple should taste sweeter than the second – you may even experience a bitter aftertaste after the second bite. So, the taste of the apple depends on the state of the palate of the person eating it, and the same apple can be perceived differently by one person.

Now let’s have a look at primary qualities. Locke states that the primary qualities of an object do not change.

Now it is your turn again.

Now ask yourself the question:

Can this distinction help us to answer sceptical arguments about the nature of the objects around us?

The argument from illusion showed that the way in which an object might appear can change even though the actual object remains the same.

For Locke, only the secondary qualities change and the primary ones stay the same. So we have reached a point where we can allow for mistaken senses, and it means that the argument from illusion is no longer a problem.

However, is there a link between the primary and the secondary qualities, and between the world and how we perceive it? Locke states:

“…the ideas of primary qualities of bodies are resemblances of them, and their patterns do really exist in the bodies themselves, but the ideas produced in us by these secondary qualities have no resemblance of them at all. There is nothing like our ideas, existing in the bodies themselves. They are, in the bodies we denominate from them, only a power to produce those sensations in us…”

(1690, Book 2, Chapter 8)

Page 50

We can say that the primary qualities are the qualities that resemble the physical object, and because of the primary qualities we can experience the secondary qualities. Therefore, although what we experience through the secondary qualities is not the same as the primary qualities, they do represent, to us, the way the world really is.

This is what we call representative realism.

Representative realism is a popular philosophical strand, and it has a version of Locke’s thesis which it calls “sense data theory”.

Sense data theory

Sense data theory postulates that we don’t immediately perceive the external object but we use sense data. Sense data are mind-dependent, and they are intermediaries in our perception. This follows Lockean secondary qualities (the secondary qualities of Locke).

According to the theory, sense data are:

1. Subjective – they are dependent upon one’s individual experience of the object

2. Existent within the mind – they forms no part of the object

3. Infallible – although it is possible to be wrong about the stick bending in water, there is no way I can be wrong about how the stick looks to me, the perceiver

4. Private – the sense data I have are particular only to me

But, we have already seen some sceptical arguments so how can we use this theory to counter them? We can shift the emphasis onto the perceiver which reduces the amount of reference we need to make to the external world; sense data seems to overcome the scepticism of Cartesian philosophy (the philosophy of Descartes in case you forgot). The sceptical arguments pointed out that the senses cannot be trusted to tell us how the world is – Descartes said that the only thing that was beyond doubt was the fact that he could think; therefore, he said he “was”.

Using the sense data theory, we know that sense data are impossible to doubt. We also know that sense data are subject to illusion, and this does help to avoid the route to scepticism.

Page 51

Also, sense data help us with the sceptical arguments that have their origin in science. This is because we know that the argument put forward by modern physics shows that there is a discrepancy between the way an object is perceived by the observer and the way that they are dissected by physics. We saw this when we looked at the tomato scenario earlier on. This shows that it is our sense data, not the tomato itself, that is giving us the picture of the red tomato in our minds.

Now look back to the attributes of sense data a few paragraphs above.

What happens if we shift all the emphasis onto the sense data and experience of the individual looking at the tomato?

The answer is that we start to lose the connection with the material world and could indeed lose it altogether. We would have a metaphysical view of the world which would put us, as the observers, into a world we are not familiar with. It would be a realm of existence that we have never had any experience of, and furthermore could never experience, because our experience would be completely different from the objects that represent the material world around us. We could go as far as saying that in fact there are no objects at all; all there is, is the experience of our perceptions.

Another sense data problem is that our view of the world would be private (point 4) and mind-dependent. This would mean that communication between two people observing the same object would be impossible, because their experiences would be so different that they would be unable to communicate their own experience. For example, if two people took a bite out of that apple we used earlier on, they would not be able to talk about the apple as an object, but only of their private and individual experience of it. They could not even communicate the point that the apple tasted sweet, because there would be no common experience between them of what “sweetness” was. So, to communicate we need to be more grounded in a public, rather than a private experience of the world around us. This is a problem we will look at a little further when we get to the section on idealism.

Page 52

Here is the answer to the activity. I think it is a good example of why it might not be a good idea to rely on individual sense data. You might disagree with me.

http://osclinks.com/4798

Some problems for sense data

1. The ontological question. If sense data have no physical existence, why do they have physical origins? Philosophers such as Russell argued that they are mental phenomena that arise in the mind, but this does not hold true. For if sense data are just from the mind, how can they mediate between our mental and physical worlds?

2. Sceptical questions. If we cannot quantify the existence of sense data and the existence of the external world, then we start to run into an argument proposed by Berkeley (we will come to him in more detail later on). If our sense data are private to each person, how is it that we see the same thing and can name the same colour – for example, we might see a brown table or a red ball and we would all know what these two things were.

3. The question of representation. This asks how we know whether sense data are representations of the real world or not. Think about geography and how you learn to read a map, because someone told you how to do it. You know what contours represent and you know how to interpret them from the lesson that you had with your teacher. But, if you saw a map for the first time, you would not understand it, and this represents how we encounter sense data. They are private, and therefore there is no teacher who will teach me how to represent them, and I do not have any direct access to what they are meant to be representing.

4. The argument from science. We have already looked at this argument when we looked at the example of the tomato. However, we can say that just because we can reduce the world to the atomic level nothing tells us that we have to do this. We can look at the world from the gross observations that we make and we can look at it from the atomic and even sub-atomic level or we can look at it from some other level in between. It is entirely up to us individually.

Page 53

Locke argues that we have a lack of choice when it comes to experiences, because they are never voluntary. This is because our perception leads to sense data in our minds, and this sense data is unavoidable.

But if we shut our eyes, we can imagine all sorts of different things. Therefore, perception is produced in our minds by an external cause; there is no choice, the sense data just appear there. Perceptions are then confirmed by the senses, but the perceptions can also be confused by the senses (we saw something similar when we looked at the act of dreaming). We could sometimes confirm our senses, for example if we decided to run a finger down the edge of a sharp knife we would bleed.

When I started this document, I started out with a blank sheet of virtual paper. I did not know what the pages would look like before I started to type, but now I have typed the words if I put it into PDF format, the page will look pretty much the same, but I will not be able to turn that page back into a blank sheet of Word document “paper”. Similarly, when you read the text, if you decided to read it out loud I wouldn’t know, but the words would exist outside of my mind, somewhere else in the physical or non-physical world.

Objections to Locke’s theory

1. There is no proof – saying there “must be something” is not the same as showing that there is by defining it.

2. There is no reasoning included to demonstrate just what sense data are.

3. Locke adds to his original argument by adding more stuff that, in turn, also needs explaining.

4. How come imagination and memory (possibly, I would argue, for memory as people can have false memories) are under our control but not sense data?

5. Why do we always get the same data from different senses?

6. There seems to be a complex interplay at work between action and perception.

We know that we experience some objects directly, so to argue this by use of direct realism can expose some weaknesses in the theory. For over 300 years philosophers have argued about this puzzle – maybe it is just a case that nobody has as yet hit on the right hypothesis. However, this theory does seem to lend itself to scepticism, and it is not an intuitive argument. More on this later.

Page 54