Analytic Tradition Lecture Notes

Spring 1995

Ben Dugan

Descartes 1596-1650

1. any two things that have different real properties are different
2. physical objects have spatial features
3. minds do not
4. minds are distinct from bodies

Knowledge of the world depends on our ideas resembling the objects that caused those ideas/mental states. God guarantees this correspondence.

Truth consists of minds coming into agreement with objects.

Hume 1711-1776

We can only access ideas -> no guarantee that the outside world or even other minds actually exist.

Examples:

• Generalization: how can we guarantee that the sun will rise tomorrow? it is irrational to make this generalization.
• Causation: This is a ridiculous conclusion. there is only correspondence, no causation.

Kant 1724-1804

Kant tried the opposite approach: minds and bodies are not distinct (truth = bodies agreeing with minds)

Knowledge is possible because minds are not passive (ala Huma and D), but rather are "wired" to construct experience in a certain manner.

A prioris: organizing activities, categories:

• substance-property
• causality
• time and space

Implications:

1. we know what we put into nature (categories/organizing activities)
2. we can know what follows from experience
3. what about that we don't experience? we know nothing.

Important to Kant is that there is no real distinction between belief and knowledge

Reactions to Kant:

1. realism (Moore)
2. logic -> Frege/Russell -> Logical Positivists
3. Phenomonology (Husserl -> Heidegger)

Logical positivism (Vienna, 1920s)

Empiricism

Knowledge is limited to experience.

Science

Only by use of the scientific method do we obtain accurate information about the world.

Logical

Relied upon development of powerful logics (Frege, etc)

Positivist

Eliminate the transcendental, supernatural, metaphysical

LP sought to use their methods to destroy all of philosophy except for logic and science.

Only sciences (esp. physics) gives us true knowledge of the world.

All problems can be solved by rational application of scientific methods

Verification principle: the meaning of a proposition is the mode of its verification. No verification = meaningless sentence (in this way, the verification principle destroys the problem of metaphysics -- renders it irrelivant, because it can't be verified). Questions like "is there an external world?" are meaningless

Verification proved to be a big thorn:

• Does verification depend on actual verification or possible verification? "There are craters on the dark side of moon."
• Complete vs. partial verification (clearly it is impossible to verify all generalizations).

1. a tautology (which makes it uninteresting)
2. verifiable ... leads to circularity

Analytic Tradition

General outlook:

1. The world is composed of "simples" or things which consist of "simples"
2. Analysis consists of decomposing complex things into their simples, which ground reality.
3. Direct experience is subjective. It should be ignored for now, but can explain eventually through careful analysis.

1. We inhabit a real world made up of objects w/ properties.
2. There are facts about the world that are context free.
3. Perception is process by which facts are registered as thoughts.
4. Thoughts can somehow cause phyisical motion.

Correspondence/picture theory of meaning:

1. Sentences are either true or false and say things about the world
2. What a sentence says (means) is a function of the words and how they are combined.
3. Words denote objects, properties, or relationships.

Wittgenstein

1. taken for granted that objects are perceived independent of language (language doesn't affect experience)
2. words name objects
3. the object named by the word is the meaning of the word (object named or the mental image named)
4. sentences are combinations of words... etc.

Wittgenstein says the picture theory certainly describes a language - a primitive system of communication - but it cannot claim to explain all of human language.

Try to use picture theory to explain words like: "not", "this", "here".

Language games -- a socially constructed system of "rules." These rules are flexible and subject to change.

We can and do create language games for special purposes. For example, see Wittgenstein's language for builders: Beam, Slab, Block

Wittgenstein's view of language, meaning, and understanding:

1. Language arises from social context.
2. Any system of signs is a language insofar as it facilitates the purpose implicit in the context in which it is used. ("Slab" is a sentence in one language and a word in another).
3. If language is effective at promoting the purpose for which it was introduced, then meaning is conveyed and understanding occurs.
4. The test of meaningfulness is not whether a language conforms to a set of criteria described by logic, but whether it is successful in its intended purpose.
5. Meaning doesn't necessarily depend upon images (slab conjures up an image to us, but to construction worker it probably doesn't).
6. "To imagine a language is to imagine a way of life."

• giving orders/obeying orders
• reporting
• acting
• singing
• hitching horses

What about universals? For picture theory, universals must exist (what object is named when we say "horse" or "square"?) With Wittgenstein, we don't need universals, because every word doesn't need to name something in particular. However, there are family resemblances between words and how we use them.

Precision: precision is not always important. In general, we get along without being very precise, or saying exactly what we mean. (This is not to say that precision is bad, however, as the Romantics did). We often know things without being able to say them.

Definitions are like rules of a game or signposts. They don't need to specify everything, nor do they guarantee anything.

A definition is exact if it is good enough.

Pelle Ehn's Alternative to the Analytic Tradition:

To participate in language games on must share the form of life within that practice is possible. This doesn't make us prisoners of our language, because we construct the games and can change the rules.

"If lions could speak, we could not understand them."

"To imagine a language is to imagine a form of life"

Language games help us understand human practice as the product of, at the same time producting the world and our understanding of it.

Note that this seems to clear up the idealist/dualist controversy idealism -> constructivism -> relativism dualism -> passive

W would see system descriptions/design artifacts as linguistic entities. they are not models/mirrors/pictures/ but rather examples, reminders, and cases

Ehn on the problems w/ systems orientation:

1. Increase division of labor and take planning out of the hands of those doing the job, rather than increase democratic labor processes.
2. Reduce the job of workers to procedures, rather than enhance skill.
3. Formalization and objectification locks the user out of the process.

Analytic approach hides direct experience, subjectivity, skill, social constructions of reality.

Practice/experience is central: We are concerned practioners first. "Practice is ontological. It is more primary than subject-object relations. Practice is acting and reflecting, but it is also social. it is produced cooperatively. To share practice is to share understanding.

Design should be oriented towards practice. ontological. -- reflective and political, looking backwards at tradition but also forwards to the as-yet-uncreated transformations.

Models aren't bad -- they support reflection and description, but they don't account for experience (prototypes and mockups). We need to know about what devices do not just how they operate.

Design is a concerned social, historical, political activity in which we anticipate computer artifacts and their use.

Use is central. (see Wittgenstein -- a language is meaningful if it is successful). We aren't just designing things, but we are changing practice, creating new forms of life.