Wednesday, February 27, 2008

Know Thyself? Part I: or, My Beef With Western Philosophy

"Know Thyself."

This famous directive, carved in the entrance-hall at the Temple of Apollo at Delphi, was a guiding philosophical principle for Socrates. The original intent of this directive is unclear. It may have meant something like, "Know your place" -- that is, know your place within your society and in relation to the gods, and don't rock the boat.

For Socrates, though, it took on a different meaning. This 'different meaning' served as one of several foundational principles in his (and, by extension, Plato's) philosophy. I think the most relevant passage in Plato is this:
I am still unable, as the Delphic inscription orders, to know myself; and it really seems to me ridiculous to look into other things before I have understood that ... Am I a beast more complicated and savage than Typhon, or am I a tamer, simpler animal with a share in a divine and gentle nature?

-- Plato, Phaedrus, 230a
Here, Socrates is explaining to Phaedrus why he has no interest in inventing rational explanations for the various Greek myths -- 'this is what really happened' sorts of stories, along the lines of 'The Red Sea really split because of an earthquake, and people thought some 'God' did it' -- something which the Sophists seem to have been involved in at the time. (More on that in a later post.)

Socrates doesn't elaborate further here on the need to "know thyself," but the meaning seems clear: How can I hope to investigate all the things around me without knowing what I -- the thing that's doing the investigating -- in fact am? How can I understand the world without first understanding the self? And this meaning is confirmed throughout Plato's philosophy, in dialogues like Phaedrus, Symposium, Republic, Meno, Phaedo...as Plato continually explores what a human being must be in order for his epistemological and metaphysical system to hold together. The ability of human beings to learn; the meaning of justice; the consequences of injustice; the whole theory of forms; in short, the whole of what is generally considered 'Platonic doctrine' -- all of it rests on the nature of the self that is built up in these dialogues. Without that self, the whole house of cards falls to the floor. With a different notion of the self, all the questions Plato asks would have to be answered differently.

This is a deep epistemological problem, and one that hasn't gone away. The history of Western philosophy is largely defined by the inability of philosophers to deal with this problem -- or to even see that it is a problem. And this holds for most of our philosophy down to the present day. While most of us have rejected what we (often mistakenly -- more on that in a later post) take to be Platonic doctrine, we're still plagued by the Platonic view of the self.

I hear members of the audience (possibly fictitious ones, since I may have no audience) screaming: "What the hell are you talking about? We're modern secular people! We've rejected dualism, the immortality of the soul, the mind as something separate from the brain; we believe the opposite of what Plato did about the self!"

To which I reply: yes, we've stripped away much of the doctrine. But only the parts we recognized as doctrine; only the parts we don't still assume are true without realizing we're making assumptions. What about the subject/object division? What about the assumption that the logic (and mental processes in general) conducted by the self can adequately describe or represent the world around us? What about the idea that the self can be treated as a concrete entity at all, even though it's constantly changing moment-to-moment, and you-ten-years-ago wouldn't be able to recognize you-now as the same entity?

I don't mean to suggest that Western philosophy has utterly failed to address questions like these. Beginning with Kant's Critique of Pure Reason, certain philosophers began to revise -- and reject -- some of our assumptions about the validity of our logic, the coherency of the self, and the general correctness of our philosophical process. Those philosophers who actually understood Kant -- few and far between (more on that in a later post) -- and their intellectual heirs have made much progress on all of these issues. In particular, I have in mind the phenomenologists and some of the existentialists (particularly the religious existentialists); but also the attacks on the stable self made by Freud and Nietzsche, and some of the work of the postmodernists. But for more than 2,000 years, our tradition was more or less hung up on the Greek notion of the self and the infallibility of the logic that arose from that notion of the self; and most of the 'serious thinking' done by intellectuals in the West still relies on these deeply flawed assumptions.

Part II later this week...

-- The Crooked Sophist

P.S.: Yes, I realize I've promised you no fewer that four(!) posts. I promise I'll write them all eventually. To keep me honest, let's keep score. I owe you:

1) Know Thyself? Part II
2) The Sophists' secular, myth-explaining agenda (which will mostly be about how Plato's philosophy is a religion)
3) What we take to be Platonic doctrine isn't really Platonic doctrine (or: Plato's a slippery, underhanded bastard and we should never take him at face value)
4) Very few philosophers understood Kant (or: Analytic philosophers don't understand what the words CRITIQUE OF PURE REASON mean)

1 comment:

io said...

In the midst of my first serious reading of Plato's dialogs, I often catch myself momentarily seduced by Socrates' neat and clean assumptions about how human beings can go about gaining knowledge:

"Consider, then, what both Hippocrates and true argument say about nature. Isn't this the way to think systematically about the nature of anything? First, we must consider whether the object regarding which we intend to become experts and capable of transmitting our expertise is simple or complex. Then, if it is simple, we must investigate its power: What things does it have
what natural power of acting upon? By what things does it have what
natural disposition to be acted upon? If on, the other hand, it takes many forms, we must enumerate them all and, as we did in the simple case, investigate how each is naturally able to act upon what and how it has a natural disposition to be acted upon by what."

Plato, Phaedrus, 270D

It's an easy and straightforward admonishment to slice the world up into piece along it's natural divisions and study the functions of the individual pieces. And the surface this method seems so obvious that only a fool or a liar would deny it.

I'm going to think of it as the entity-relationship (ER) model of knowledge acquisition. When designing a database one of the first steps is to create an ER model of the information that's going into the database. The model has three major components: entities, attributes, and relationships between entities. An entity is one of the subjects of the model (and usually corresponds to a database table); to take a page from a classic example, an ER model of a purchasing department will divide up the stream of information going through the purchasing department into 5 entities (orders, customers, sales representatives, stock, products) that will each be considered a Breed of Thing in the ER model. Each one will have a set of attributes (i.e. sales reps will have names and salaries, orders will have dates, stock will have quantities). The entities will also have relationships with one another: every order is related to at least one customer, at least one salesperson, and at least one product. Socrates systematic thinking about nature and ER models both seem straightforward and fool proof at the surface. More concretely, these seem to be objective systems: two people using the system to analyze the same situation will come up with the same results, the same model.

The models appear objective because they ask us to go out into the world and look for "natural categories" and "natural subdivisions" of categories.

However, as any DBA or metadata librarian familiar with AACR2 can tell you, when real people apply the same method for finding natural categories and subdivisions their results differ. Many a wise DBA or metadata librarian will caution students that ER modeling or assigning Library of Congress subject headings is an art rather than a science. I would put it this way: decisions about the boundaries of natural categories and the locations of natural subdivisions have a high degree of arbitrariness regardless of the method used to judge them.

If you start creating ER models for databases, you quickly realize that you often have several choices for modeling the same real world situation and all of them are equally valid. Is commission percentage an attribute of sales person or of an order or of a class of sales person (for example, top earners or those with seniority)? Does exactly one sales person work on every order, or on occasion do multiple sales people contribute to making a single order? Working in concert with the people who are going to use the database -- the people who live in the real world you're modeling -- you make decisions about how to model all the situations (including deciding which situations are relevant and which are "out of scope") according to a "system" that amounts to a helter-skelter mish-mash of aesthetic preference, technological efficiency, flip-of-coin, and bowing to other people's preferences.

I think the same must be true of Socrates' ER style of knowledge acquisition. A practitioner believes that the method is objective and will produce the same results every time, but a lot of values other than the method go into the process (i.e., culture, aesthetic, chance, input from others). The model is neat and clean in the abstract/in theory but confused and messy in the concrete/practice.

But the model sounds so good, especially from handsome Socrates, that it is difficult to remember that any theoretical or abstract description of it tends to sweep under the rug these problems:

1. What if there are no natural categories?
2. What if there are no natural subdivisions?
3. What if human beings are incapable of rightly discerning natural categories and subdivisions?
4. What if dividing the world into its natural categories and subdivisions does NOT produce meaningful knowledge?

In ER modeling for databases, you can ignore these questions because you aren't looking for Truth, but rather for Acceptance and Efficiency. A decent model of roughly how people imagine they would best like to conduct their business is sufficient; even though the people may suffer from limited imaginations and fail to dream of the Ideal way to conduct their business, they'll never be aware of what they can't even imagine and will be delighted with the database that elegantly embodies what they say they want. This is all fine and good for database modeling, but exactly the opposite of fine and good for Truth Seeking according to Socrates. The very conditions that make a successful ER model (human acceptance and efficiency) make a failed attempt at knowledge. Given this it's ironic and sad that Socrates fails to address the limitations of this type of systematic pursuit of the truth (et tu, Plato?).

I wonder how they would answer my four critiques? How do you go about resolving a debate like "are there natural categories or all all categories man made?"

I have more sympathy than ever for social constructionists now that I've read some Plato. To argue that every category is constructed by human beings in concert (think acceptance and efficiency as criteria) is to argue against Socrates and in many ways against the premises of the scientific method. To argue against natural categories and subdivisions leaves us in a fog: how can we even have debate the issue?