So much debate, so little time. Should the government regulate executive compensation? I'll ignore, for now, whether such regulation is realizable (for all kinds of reasons), consistent with the American vision, and popularly supported. Instead, I conducted a thought experiment to determine whether there would be ANY condition under which it would be appropriate to regulate executive compensation. The answer is unequivocally yes: if executive compensation so undermined, so endangered, so threatened the common good and/or the welfare of the nation that it was a clear and present danger, then it should be regulated just as any other danger would be (including economic concerns), e.g., sale of methamphetamines, child labor, disposal of radioactive waste, indentured servitude, export of weapons technology, etc.
If that is the case, then the real concern is not the size of executive compensation packages per se. (The flip side of Jesus' "the poor you will always have with you" is that the oligarchs you will also always have with you). What seems much more problematic is that compensation mechanisms may have been a proximate cause of the financial crisis and the "great recession," and indeed may be at the root of a great many other economic pressures that are not in the public interest. There is, in fact, a great deal of compelling evidence for this. Many compensation schemes we have heard about handsomely, obscenely, rewarded executives and others for precisely those actions and decisions that precipitated, is precipitating the ongoing disaster: short term metrics at the expense of long term prospects, extreme leveraging of positions, IBGYBG attitudes, hedging against one's own positions, abandoning prior obligations to employees and retirees, cronyism, cowing regulators and buying legislators, ignoring blatant conflicts of interest. If it walks like a clear and present danger and quacks like a clear and present danger...
Monday, January 25, 2010
Thursday, January 7, 2010
Does a Bear ...
Ooh, ooh, what's it called, when you "complicate" a well known saying? Kind of the reverse of a euphemism, but not quite. One occurred to me as I was putting together the last post - and got me all agiggle:
Does a bear increase the overall entropy of the universe while converting energy from one form to another in the woods?
On second thought, it's too obvious - surely that saying or a close variant is already out there, right?
Update: circumlocution or periphrasis? Not quite - they don't seem to carry the intentional humorous connotation that is always an element in these sayings.
Update: Proposed T-shirt for physics students:
On the front: "Do I like physics?"
On the back: "Does a bear increase the overall entropy of the universe while converting energy from one form to another in the woods?"
Update: Proposed tee shirt for chemists, engineers, biologists, oh, hell, anybody:
On the front: "Do I like thermodynamics?"
On the back: "Does a bear increase the overall entropy of the universe while converting energy from one form to another in the woods?"
Does a bear increase the overall entropy of the universe while converting energy from one form to another in the woods?
On second thought, it's too obvious - surely that saying or a close variant is already out there, right?
Update: circumlocution or periphrasis? Not quite - they don't seem to carry the intentional humorous connotation that is always an element in these sayings.
Update: Proposed T-shirt for physics students:
On the front: "Do I like physics?"
On the back: "Does a bear increase the overall entropy of the universe while converting energy from one form to another in the woods?"
Update: Proposed tee shirt for chemists, engineers, biologists, oh, hell, anybody:
On the front: "Do I like thermodynamics?"
On the back: "Does a bear increase the overall entropy of the universe while converting energy from one form to another in the woods?"
Testing the Emergence Hypothesis
In considering the hypothesis of emergent phenomena there is, from a science perspective, a need and a responsibility to test the hypothesis, and so a singularly important question to be answered: what are the distinctive observable characteristics that an emergent phenomenon would display compared to one that is reductive. As I have already pointed out in this blog, there is a perfectly lovely historical example of such an approach in Einstein’s development of the Special Theory of Relativity. As Einstein exemplified in that case, it is one of the responsibilities of a scientist advancing a novel hypothesis to identify how it may be tested.
Now I suspect that some with a weaker grasp of the principles and practice of science would propose a straightforward test of demonstrating that an emergent principle correctly predicts a phenomenon where reductive principles were incapable of making the prediction. But I think most practicing scientists would recognize that this is not a sufficient test of the emergence hypothesis. Reductive principles may still correctly characterize a complex phenomenon in spite of the fact that we do not have the capability to apply them to predict it. Let’s take an extreme case to make the point: our capability to “predict” phenomena with quantum field theory is based on an infinite perturbation expansion – that is, everything, literally everything you can actually “predict” with it, is intrinsically an approximation; and yet no emergence proponents, not a one, has ventured to claim that the magnetic dipole moment of the electron is an emergent phenomenon even though we know we do not have the capability to predict it exactly. (Some folks just seem to go all wobbly over agreement to one part in ten billion.) So, let’s recap: even if a hypothesized emergent principle correctly predicted a phenomenon that reductive principles couldn’t because of our capability to apply them, this is not, I repeat not, a sufficient test of the emergence hypothesis.
Because of the capability dilemma, we are forced to apply a weaker and admittedly much less satisfying test of reductive principles for complex phenomena: examining whether the phenomena are consistent with the reductive principles, that is, examining a complex phenomenon looking for evidence that it violates the reductive principles. (Giraffe composed of atoms from the periodic table of the elements – check. Giraffe accelerates at 9.8 m/s^2 when dropped – check. Giraffe increases overall entropy of the universe when converting energy from one form to another – check. Etc., etc., etc.) Of every test of this kind of which I am aware (save one) there has never been a single instance that provided even the remotest evidence of violation of reductive principles. (The one instance that I am genuinely puzzled by is “dark energy”, which, oddly, seems to draw little attention from emergence proponents. That dark energy is some creepy weird shit, man!) And so here is the kicker for the emergence proponents: If a phenomenon is correctly described by an emergence hypothesis and yet the phenomenon is still consistent with reductive principles, then Occam’s Razor strongly suggests, if not demands, that the emergence hypothesis be considered unnecessary.
So, a challenge to emergence proponents: clearly and unequivocally identify the distinctive observable characteristics that an emergent phenomenon would display compared to one that is reductive and in a way that satisfies Occam. If you cannot or will not do this then it suggests that you are not doing science.
Update: OK, for the purpose of the challenge, let’s say reductive phenomena are those that are solely the consequence of the standard model of particle physics + the general theory of relativity + any as yet undiscovered components of the universe consistent with these [Higgs boson, dark matter (? – see NPR 13.7 Blog post @Mgleiser: Dark Matter), dark energy(?!?)] + any underlying reductive structure that subsumes these. Yuk. Not so pretty. But that’s still part of the problem: a claim of emergence could appear to offer clarity simply in contrast to murkiness in parts of our reductive understanding.
I know someone is going to call foul on this because we don’t know what the undiscovered components are and what the underlying reductive structure is. But turnabout is fair play: it’s just as much of a foul for emergence proponents to exclude these, because there are a number of very strong indicators that they are there. Without a knowledge of the undiscovered components and understanding of the underlying reductive structure a claim that a phenomenon is emergent cannot genuinely be tested against a competing reductionist claim, and as I have argued, if a phenomenon can be understood in terms of both an emergent hypothesis and a reductive hypothesis, Occam’s razor says reductive wins.
The posts on the NPR 13.7 blog suggest a range of interpretations of what constitutes emergence. For the purpose of the challenge, let’s say emergence is behavior which is demonstrably not solely a consequence of reductive phenomena as defined above.
Update: I hadn't noticed before I wrote this, but in the comments to one of the NPR 13.7 blog posts (see @SAK42: Breaking the Galilean Spell), there is indeed some heady discussion of a possible (extraordinarily speculative)relationship between dark energy and emergence. Like I said, that dark energy ...
Now I suspect that some with a weaker grasp of the principles and practice of science would propose a straightforward test of demonstrating that an emergent principle correctly predicts a phenomenon where reductive principles were incapable of making the prediction. But I think most practicing scientists would recognize that this is not a sufficient test of the emergence hypothesis. Reductive principles may still correctly characterize a complex phenomenon in spite of the fact that we do not have the capability to apply them to predict it. Let’s take an extreme case to make the point: our capability to “predict” phenomena with quantum field theory is based on an infinite perturbation expansion – that is, everything, literally everything you can actually “predict” with it, is intrinsically an approximation; and yet no emergence proponents, not a one, has ventured to claim that the magnetic dipole moment of the electron is an emergent phenomenon even though we know we do not have the capability to predict it exactly. (Some folks just seem to go all wobbly over agreement to one part in ten billion.) So, let’s recap: even if a hypothesized emergent principle correctly predicted a phenomenon that reductive principles couldn’t because of our capability to apply them, this is not, I repeat not, a sufficient test of the emergence hypothesis.
Because of the capability dilemma, we are forced to apply a weaker and admittedly much less satisfying test of reductive principles for complex phenomena: examining whether the phenomena are consistent with the reductive principles, that is, examining a complex phenomenon looking for evidence that it violates the reductive principles. (Giraffe composed of atoms from the periodic table of the elements – check. Giraffe accelerates at 9.8 m/s^2 when dropped – check. Giraffe increases overall entropy of the universe when converting energy from one form to another – check. Etc., etc., etc.) Of every test of this kind of which I am aware (save one) there has never been a single instance that provided even the remotest evidence of violation of reductive principles. (The one instance that I am genuinely puzzled by is “dark energy”, which, oddly, seems to draw little attention from emergence proponents. That dark energy is some creepy weird shit, man!) And so here is the kicker for the emergence proponents: If a phenomenon is correctly described by an emergence hypothesis and yet the phenomenon is still consistent with reductive principles, then Occam’s Razor strongly suggests, if not demands, that the emergence hypothesis be considered unnecessary.
So, a challenge to emergence proponents: clearly and unequivocally identify the distinctive observable characteristics that an emergent phenomenon would display compared to one that is reductive and in a way that satisfies Occam. If you cannot or will not do this then it suggests that you are not doing science.
Update: OK, for the purpose of the challenge, let’s say reductive phenomena are those that are solely the consequence of the standard model of particle physics + the general theory of relativity + any as yet undiscovered components of the universe consistent with these [Higgs boson, dark matter (? – see NPR 13.7 Blog post @Mgleiser: Dark Matter), dark energy(?!?)] + any underlying reductive structure that subsumes these. Yuk. Not so pretty. But that’s still part of the problem: a claim of emergence could appear to offer clarity simply in contrast to murkiness in parts of our reductive understanding.
I know someone is going to call foul on this because we don’t know what the undiscovered components are and what the underlying reductive structure is. But turnabout is fair play: it’s just as much of a foul for emergence proponents to exclude these, because there are a number of very strong indicators that they are there. Without a knowledge of the undiscovered components and understanding of the underlying reductive structure a claim that a phenomenon is emergent cannot genuinely be tested against a competing reductionist claim, and as I have argued, if a phenomenon can be understood in terms of both an emergent hypothesis and a reductive hypothesis, Occam’s razor says reductive wins.
The posts on the NPR 13.7 blog suggest a range of interpretations of what constitutes emergence. For the purpose of the challenge, let’s say emergence is behavior which is demonstrably not solely a consequence of reductive phenomena as defined above.
Update: I hadn't noticed before I wrote this, but in the comments to one of the NPR 13.7 blog posts (see @SAK42: Breaking the Galilean Spell), there is indeed some heady discussion of a possible (extraordinarily speculative)relationship between dark energy and emergence. Like I said, that dark energy ...
Saturday, January 2, 2010
Dr. Kauffman’s Challenge
The discussion between Dr Kauffman and Dr. Goodenough seems to me to pivot on what is admissible as natural law. Dr. Kauffman, at least in my reading, appears to be advancing a position – really, a postulate – that in order for a principle to be admissible as a natural law all, phenomena associated with it must be deducible, e.g., for the principle of evolution to be admissible as a natural law one must, at least in principle, be able to deduce the giraffe – no deducible giraffe, “no law”. Dr. Goodenough, though, is advancing what in my experience is the much more common view within science, one that has proven at least empirically successful, that a principle is admissible as natural law if some phenomena are deducible from it and the remainder of the relevant phenomena are at least consistent with it, e.g., evolution is admissible as a natural law because from it one can deduce that some species that are subject to a rapid change in their environment will suffer extinction, and because giraffes are consistent with evolution.
Because of the more commonly accepted and empirically successful understanding, I would suggest that the onus is genuinely on Dr. Kauffman to do more than ask us to simply accept as given his postulate about what principles are admissible as natural law. Finally, I claim that this is not an insignificant point: we have an historical/scientific precedent that bears on this issue so, even though I deeply apologize for the apparently tangential nature and length of the following paragraph, please humor me.
In my experience most people very seriously misunderstand the true nature of what Einstein proposed in the special theory of relativity (STR). It is the first postulate, considered so unglamorous that it is rarely even mentioned in popular treatments of STR, that is perhaps its most profound insight. The first postulate of the STR was, in fact, a postulate about what principles are admissible as physical law: principles that take the same mathematical form in all inertial reference frames. At the time Einstein was developing the theory, there were two different mathematical representations of a single electrodynamic phenomenon, and which representation was used depended on the inertial reference frame in which the phenomenon occurred. Both representations worked flawlessly in experiments, but Einstein’s key insight, the reason he was unwilling to let this status quo stand, the burr under his saddle if you will, was what principles he considered admissible as physical law. He found that the two different representations could be shown to have the same mathematical form in any inertial reference frame provided one accepted the painfully counterintuitive assumption that the speed of electomagnetic radiation (light) is the same in all inertial reference frames. Thus, the second, sexier, postulate of the STR is, in some sense, simply a deduction following from the first postulate and the flawless experimental results associated with the two electrodynamic principles. From this point, Einstein went on to predict a number of consequences (e.g., E=mc^2) in terms of the specific observable (and, again, often painfully counterintuitive) behavior of physical systems that contrasted with the competing view. The point (yes I do have a point) is that tests of the special theory of relativity, such as experiments investigating the constancy of the speed of light and the equivalence of mass and energy were, in fact, tests of Einstein’s postulate about what principles are admissible as physical law against a competing understanding.
Dr. Kauffman’s challenge: Using the historical/scientific precedent of the STR model outlined above, if you are proposing a postulate about what principles are admissible as natural law, as it clearly seems to me you are, a postulate that contrasts with a commonly used and empirically successful understanding, I maintain that it is incumbent upon you as a practicing scientist to identify what the consequences of that postulate would be in terms of the specific observable behavior of natural systems that constrasts with the current understanding and, further, to propose a realizable test of the postulate. In conclusion it is, I suspect, pertinent to this dicussion, that a part of the reason I am so sensitive on this topic is that a fatal flaw of ID “theory” is its failure, its unwillingness really, to meet these minimal requirements.
Because of the more commonly accepted and empirically successful understanding, I would suggest that the onus is genuinely on Dr. Kauffman to do more than ask us to simply accept as given his postulate about what principles are admissible as natural law. Finally, I claim that this is not an insignificant point: we have an historical/scientific precedent that bears on this issue so, even though I deeply apologize for the apparently tangential nature and length of the following paragraph, please humor me.
In my experience most people very seriously misunderstand the true nature of what Einstein proposed in the special theory of relativity (STR). It is the first postulate, considered so unglamorous that it is rarely even mentioned in popular treatments of STR, that is perhaps its most profound insight. The first postulate of the STR was, in fact, a postulate about what principles are admissible as physical law: principles that take the same mathematical form in all inertial reference frames. At the time Einstein was developing the theory, there were two different mathematical representations of a single electrodynamic phenomenon, and which representation was used depended on the inertial reference frame in which the phenomenon occurred. Both representations worked flawlessly in experiments, but Einstein’s key insight, the reason he was unwilling to let this status quo stand, the burr under his saddle if you will, was what principles he considered admissible as physical law. He found that the two different representations could be shown to have the same mathematical form in any inertial reference frame provided one accepted the painfully counterintuitive assumption that the speed of electomagnetic radiation (light) is the same in all inertial reference frames. Thus, the second, sexier, postulate of the STR is, in some sense, simply a deduction following from the first postulate and the flawless experimental results associated with the two electrodynamic principles. From this point, Einstein went on to predict a number of consequences (e.g., E=mc^2) in terms of the specific observable (and, again, often painfully counterintuitive) behavior of physical systems that contrasted with the competing view. The point (yes I do have a point) is that tests of the special theory of relativity, such as experiments investigating the constancy of the speed of light and the equivalence of mass and energy were, in fact, tests of Einstein’s postulate about what principles are admissible as physical law against a competing understanding.
Dr. Kauffman’s challenge: Using the historical/scientific precedent of the STR model outlined above, if you are proposing a postulate about what principles are admissible as natural law, as it clearly seems to me you are, a postulate that contrasts with a commonly used and empirically successful understanding, I maintain that it is incumbent upon you as a practicing scientist to identify what the consequences of that postulate would be in terms of the specific observable behavior of natural systems that constrasts with the current understanding and, further, to propose a realizable test of the postulate. In conclusion it is, I suspect, pertinent to this dicussion, that a part of the reason I am so sensitive on this topic is that a fatal flaw of ID “theory” is its failure, its unwillingness really, to meet these minimal requirements.
Subscribe to:
Posts (Atom)
