June 20, 2007
FAS 5, Lawyer's Intuitions, Algorithms, and the Goldenlocks Mean
Posted by Jeff Lipshaw
I have posted before on the concept I call "the ontology of FAS 5." FAS 5, for the uninitiated, is the standard under generally accepted accounting principles, or GAAP, for loss contingencies, the quantification or description of something bad that may happen, but hasn't yet, like a liability claim or a lawsuit. I keep returning to what I think is one of the most intriguing sentences in all of of lawyerdom: the explanation in the commentary to the ABA's Statement of Policy regarding FAS 5.
Lawyers do not generally quantify for clients the “odds” in numerical terms; if they do, the quantification is generally only undertaken in an effort to make meaningful, for limited purposes, a whole host of judgmental factors applicable at a particular time, without any intention to depict “probability” in any statistical, scientific or empirically-grounded sense.
My reading this summer consists of a dive into the philosophy of science and social science. (The books and articles are stacked up here in my office, and scold me every time I walk by them to the computer.) The idea that the concept at the very heart of loss contingency, which is at the very heart of the model by which we represent a business, is not "statistical, scientific or empirically grounded" is fascinating to me. So what is the sum of those judgmental factors, and is the import here that the lawyer's views offered as the basis for an accounting entry (or, more likely, non-entry) are intuitive, non-scientific, and based in a priori knowledge?
I just finished reading Douglas Hofstadter's I Am a Strange Loop (sorry if I keep repeating this). He did a great service by writing the book in the order he did. The thoughts at the beginning, on how we deal with complexity, and how complex systems take on the appearance (if not the reality) of consciousness, have rational force. (The idea that a stack of books is complex enough to scold me is only metaphor, however, and not a statement on my part that they are conscious.) The arguments at the end, to the effect that there can be no transcendental or spiritual aspect to consciousness, because all can be explained by physics, are less compelling, and, indeed, seem tautological to me. In his epilogue, he refers to Godel's theorem, and its impact on the completeness (versus consistency) of Principia Mathematica as the "Trojan horse that sneaked self-consciousness into the very fortress that was built to keep it out. . . ." No! There's no evidence whatsoever that arithmetic is self-conscious! There is evidence that arithmetic is so complex that it develops propositions that are undecidable under its own meta-principles. It is a demonstration of limits; not in my mind [sorry!], a debunking of human consciousness into mere cause-and-effect empirical reality. (As I said, I'm agnostic on the subject, which means that I'm inclined to disagree with Professor Hofstadter, but I'm open to better evidence or argument than he presents.)
That interplay of limits and complexity, of the paradoxes built into everything, including the fact that we can articulate the concept of the unknowable or impossible, even when we can't know what the unknowable or impossible is, seems like it would be unduly metaphysical, but my point is that it keeps popping up in the midst of things as this-world and pragmatic as lawyers' responses to loss contingency issues under FAS 5.
Related to our perceptual limits is the perceptual editing function of a human (or, I suspect, a non-human) mind. It's rational that natural selection would have favored those beings that could distinguish the data important to survival. We perceive on a limited band-width, to use the jargon. But that perception is, as of yet, non-algorithmic. John D. Barrow, Professor of Mathematical Sciences and Director of the Millennium Mathematics Project at Cambridge, says in his book, Impossibility: The Limits of Science and the Science of Limits, "Remarkably, no computer has yet managed to reproduce our many levels of visual sensitivity to patterns." All unduly metaphysical, you think, until you ponder the "I know it when I see it" aspect (as I am now in early prep for securities regulation) of materiality. It's the Goldilocks test: not too much information, not too little information, but information that is juuuust right.
Aretaic philosophers, like my friend Larry Solum (Illinois, above left, oops no, that's Aristotle), model this nicely within the Goldenlocks Mean, and Aristotle's metaphysics are probably a good working basis for corporate and securities lawyers who feel the philosophical urge (not recommended when responding to SEC staff comments on an S-1 or S-4 or during a closing, but okay in the bar afterwards). I'd take the Golden Mean (with its implicit intuitiveness) over algorithm any day. Just a little more from Professor Barrow: "There is an intriguing pattern to many areas of deep human inquiry. Observations of the world are made; patterns are discerned and described by mathematical formulae. The formulae predict more and more of what is seen, and our confidence in their explanatory and predictive power grows. . . . Users of the magic formulae begin to argue that they will allow us to understand everything. The end of some branch of human inquiry seems to be in sight. . . ." Well, for the punch line to this, pick up Professor Barrow's book.
TrackBack URL for this entry:
Listed below are links to weblogs that reference FAS 5, Lawyer's Intuitions, Algorithms, and the Goldenlocks Mean: