The Anthropic Principle
The anthropic principle seems to have lately touched a raw nerve in the physics community. Prominent physicists have come out with uncharacteristically emotional statements such as "Anyone who talks about the anthropic principle is a scoundrel", "I hate the anthropic principle", and "it smells of religion and intelligent design"(1). On the other hand, equally prominent physicists have theorized "We might have to accept the idea that the existence of life was built into the laws of physics from the beginning".
The anthropic principle seems prima facie to support the ideas of those who have a strong faith in the existence of an external intelligent designer of the universe. On the other hand there is an interpretation seemingly supportive of those who have an equally unshakeable faith in the absence of such an intelligent designer. We label each of these belief systems as "faiths" because neither can be rigorously established by scientific methods. Either faith seems, by those in the opposite camp, to have bizarre consequences.
I believe,however, that the principle can be stated in a neutral, non-confrontational, and scientifically quantifiable formulation:
One might like to replace the word "pondering" by "understanding" but that is more speculative and certainly premature at least. Some might like to replace the word "pondering" by "appreciating" but this brings in an aesthetic sense that, while possibly intrinsic to the goal of physics, is stronger than necessary for present purposes.
Apart from the modifying phrase "within narrow bounds" the statement is obvious and non-controversial. Exactly how narrow the bounds are, outside of which no currently imaginable life forms could exist, can be and has been the subject of scientific investigation as, for example in (2). In some cases the bounds are extremely narrow. For instance, on the basis of the anthropic principle, Weinberg (3) predicted that Einstein's cosmological constant could be no more than 10^(-120) in natural units. The recent experiments suggesting a non-zero value of this constant fit comfortably within Weinberg's prediction.
An even earlier result was the prediction of Fred Hoyle et al.(4) that, for nucleosynthesis to produce elements as heavy or heavier than Carbon, which seem essential to life as we know it or can consistently imagine, a previously unknown resonant state of Carbon needed to exist and to have a narrowly defined mass. Such a resonant state with the requisite mass was later discovered to exist. A recent analysis (6) of several nuclear models suggests that, if the nuclear force differed from its observed strength by as much as two tenths of a percent the heavy elements essential to life would not have been produced in stellar nucleosynthesis. A host of other "anthropic coincidences" have been catalogued in the book by Barrow and Tipler(5).
In spite of these results, the anthropic principle has been criticised as "vague" or "unpredictive" by some authors. That is, in some cases, it is difficult to tell where the boundary is between a physics that is hospitable to life and one in which no conceivable life forms could arise. Perhaps we have been too unimaginative in conceiving of alternate life forms. However, few would argue for the possibility of matter-based intelligent life in a world with no stable atoms or in a world consisting only of hydrogen and helium as might exist if the neutron-proton mass difference were only slightly different from its observed value. Perhaps also, as some have claimed, there are alternative physics scenarios with a viable biology that are nevertheless not realized in nature. This, however, is not contrary to the anthropic principle as formulated above which does not necessarily claim that the observed universe is uniquely determined by anthropic selection.
It has also been charged that the anthropic principle is unscientific. This may or may not be true depending on one's definitions of "scientific". On the other hand it might be that there are realms of knowledge that cannot be probed by traditional scientific methods. We could, for example, adopt the narrow definition:
Physics is the science of all that can be established by experimentation on, or observation of, inanimate objects without reference to causes outside the universe.
This definition restricts the scope of physics without prejudice as to whether there are external causes or interactions of mind and matter or even whether "mind" exists apart from matter. It does, however, label as unscientific any hypotheses involving universes that are intrinsically, causally disjoint from ours. In any case we should avoid implying any pejorative connotations to the term "unscientific". Also, although the proper subject matter of physics may be "inanimate", one can note that the presence of an "observer" is an essential feature of quantum mechanics and that animate objects also obey the laws of physics.
Because physics is the most mathematically precise of all the sciences, physicists sometimes imply that only their methods can provide answers as to why things are as they are. In fact, it is clear that physical reasoning provides only connections between various principles and phenomena and never provides a final and fundamental answer to the question "why". Every solution in physics comes together with further questions. Consider the following typical train of physical reasoning:
(Q) Why does the moon revolve around the earth and the earth around the sun?
(A) Because of Newton's universal inverse square law of attraction.
(Q) But why is the law "inverse square"?
(A) Because the graviton is a massless particle.
(Q) But why is the graviton massless?
(A) Because all the forces of nature are gauge interactions.
(Q) But why are the forces gauge forces?
(A) Because gauge forces emerge naturally from string theory.
(Q) But why?... but why?... but why?...
Each "answer" above summarizes a possible physics seminar or even a semester-long course. Often such a train of questions can only be ended by the statement that, if things were different, we would not be here to discuss the question. This is the point at which the anthropic principle is currently entering persistently into physics discussions. By offering such an answer, perhaps we are going beyond physics into metaphysics. Perhaps this is a sign of the maturity of physics at this point in the 21st century although no one should propose that there are no further connections to be made by traditional physics methods. Perhaps we are at the point where much might be learned by expanding the above definition of physics or by defining a new and broader science. In any case, the fact that chains of questions such as the above have no final answer within physics, as currently defined, should not blind us to the fact that, in the process, we have uncovered an amazing elegance in the natural laws.
The anthropic principle, if accepted as an intrinsic property of the material universe, does provide a mode of response to questions that are beyond the current capabilities of theoretical physics or even beyond the definition of what questions physics can address. It is possible to look at each possible question about nature from the point of view of the principle. For instance, among many others, the questions of dark matter and of neutrino masses have been investigated from this point of view (7). The question of why our universe appears to have three space dimensions came under physics scrutiny with the rise of string theory but is still without any definitive, scientific answer. From the anthropic point of view, however, it has been noticed that in a universe with more than three dimensions of the nature of x, y, and z, there would be no stable planetary orbits thus precluding the evolution of life. If there were fewer than three, it has been argued that the highly connected neural networks in the human brain would not have been possible (8). Finally one can note that there are questions that have still never been asked from the scientific point of view. One of these is why nature is quantum mechanical at all rather than classical. From the anthropic point of view the answer is obvious: in classical physics the atom is unstable thus disallowing the evolution of stars, planets, and life while in quantum mechanics the atom is stable or at least much longer lived than the universe todate.
From the dawn of physics in ancient Greece up to modern times there has been a competition between two threads of thought: intelligent design vs. extreme randomness. One, championed by Anaxagoras and Aristotle holds that nothing happens without an efficient (mechanical) cause and a final cause (purpose). The ultimate cause was labeled "the Mind". The other, championed by Leucippus and Democritos, holds that everything happens by "chance" in a universe of random particle collisions. "In reality there is nothing but atoms and the void." The former point of view was dominant from the time of Aristotle until at least the advent of quantum mechanics in the 1920's. It seemed to lend itself to a theistic worldview and, indeed, it was greatly elaborated by the medieval philosophers. Quantum mechanics forced the abandonment of the notion that specific events at the atomic level have precise causes within the physical systems themselves. The laws of physics since the birth of quantum theory have been statistical in nature rather than deterministic. In reducing physical law to probability theory, it was considered natural by some to abandon notions of meaning and purpose in nature. This view seems to lend itself to a non-theistic weltanschauung but, if one insists on a causal theory, it has also a theistic interpretation involving causal agents external to the physical system.
Much of the hostility that the anthropic principle has met is due to its possible suggestion of a deep purpose built into the laws of physics and driving the evolution of the material universe. However, just as, since Galileo, believers in religion need to accommodate the truths derived from scientific experimentation, believers in the absence of an external designer might need to accommodate a fundamental principle building into the laws of physics the eventual evolution of intelligent life. Several imaginative proposals have been advanced to reconcile the anthropic principle with the assumed absence of fundamental causes or purposes.
The term "anthropic principle" was coined by the astrophysicist B. Carter (9). It can be discussed in either a "weak form" or a "strong form" which can be paraphrased as follows:
We use the term universe to describe a region of space-time with a common set of physical constants, elementary particles, and fundamental forces. Both the strong form and the weak form have theistic as well as non-theistic interpretations.
The strong form claims to answer Einstein's famous question as to whether nature had a choice in setting down the laws of physics. It suggests, remarkably, that the only possible universe is the one (ours) which allows the evolution of life. It appeals to those who seek to find a unique, mathematically consistent string theory that will "explain" why things are as we see them. However, the strength of science, namely its requirement of confirmation by reproducible experiments, is also its greatest weakness namely its contingency. No matter how "unique" and compelling a theory might be, it still requires experimental confirmation which shows that no physical theory is realized necessarily in nature. If physics ends with a final string theory of everything as some hypothesize, one would still have to ask why this theory is realized rather than, say, nothing or one of the free (non-interacting) field theories which are at present the only non-perturbatively soluble mathematically consistent theories in our four dimensional space.
The weak form appeals to those who believe in ultimate randomness with no purpose or fundamental causes in nature. If some measured value, such as the cosmological constant discussed above, takes an anthropically selected value of the order of 1/n of its natural value, this might indicate the existence of at least n alternative universes which are incompatible with the evolution of life. If nature makes random quantum jumps between these alternative universes, eventually it will land in one of the few where life is possible and then life will evolve as it has on earth. If there are many "anthropic coincidences" each with an a priori probability of 1/ni, the number of alternative universes is, at least, the product of the ni. Supposedly, some string theorists are attempting to enumerate the local ground states of string theory. The cosmological constant "coincidence" suggests that there should be at least 10120 local minima in string theory. It is not clear whether this theory of a vast number of alternative universes where life is not possible is experimentally verifiable as required if it is to be a "scientific" theory in the current sense of the word. Perhaps, instead, it is part of a coming new metaphysics.
The anthropic principle seems prima facie to support the ideas of those who have a strong faith in the existence of an external intelligent designer of the universe. On the other hand there is an interpretation seemingly supportive of those who have an equally unshakeable faith in the absence of such an intelligent designer. We label each of these belief systems as "faiths" because neither can be rigorously established by scientific methods. Either faith seems, by those in the opposite camp, to have bizarre consequences.
I believe,however, that the principle can be stated in a neutral, non-confrontational, and scientifically quantifiable formulation:
- The anthropic principle is the observation that, within narrow bounds, the laws of physics are such as to allow the evolution of creatures capable of pondering the laws of physics.
One might like to replace the word "pondering" by "understanding" but that is more speculative and certainly premature at least. Some might like to replace the word "pondering" by "appreciating" but this brings in an aesthetic sense that, while possibly intrinsic to the goal of physics, is stronger than necessary for present purposes.
Apart from the modifying phrase "within narrow bounds" the statement is obvious and non-controversial. Exactly how narrow the bounds are, outside of which no currently imaginable life forms could exist, can be and has been the subject of scientific investigation as, for example in (2). In some cases the bounds are extremely narrow. For instance, on the basis of the anthropic principle, Weinberg (3) predicted that Einstein's cosmological constant could be no more than 10^(-120) in natural units. The recent experiments suggesting a non-zero value of this constant fit comfortably within Weinberg's prediction.
An even earlier result was the prediction of Fred Hoyle et al.(4) that, for nucleosynthesis to produce elements as heavy or heavier than Carbon, which seem essential to life as we know it or can consistently imagine, a previously unknown resonant state of Carbon needed to exist and to have a narrowly defined mass. Such a resonant state with the requisite mass was later discovered to exist. A recent analysis (6) of several nuclear models suggests that, if the nuclear force differed from its observed strength by as much as two tenths of a percent the heavy elements essential to life would not have been produced in stellar nucleosynthesis. A host of other "anthropic coincidences" have been catalogued in the book by Barrow and Tipler(5).
In spite of these results, the anthropic principle has been criticised as "vague" or "unpredictive" by some authors. That is, in some cases, it is difficult to tell where the boundary is between a physics that is hospitable to life and one in which no conceivable life forms could arise. Perhaps we have been too unimaginative in conceiving of alternate life forms. However, few would argue for the possibility of matter-based intelligent life in a world with no stable atoms or in a world consisting only of hydrogen and helium as might exist if the neutron-proton mass difference were only slightly different from its observed value. Perhaps also, as some have claimed, there are alternative physics scenarios with a viable biology that are nevertheless not realized in nature. This, however, is not contrary to the anthropic principle as formulated above which does not necessarily claim that the observed universe is uniquely determined by anthropic selection.
It has also been charged that the anthropic principle is unscientific. This may or may not be true depending on one's definitions of "scientific". On the other hand it might be that there are realms of knowledge that cannot be probed by traditional scientific methods. We could, for example, adopt the narrow definition:
Physics is the science of all that can be established by experimentation on, or observation of, inanimate objects without reference to causes outside the universe.
This definition restricts the scope of physics without prejudice as to whether there are external causes or interactions of mind and matter or even whether "mind" exists apart from matter. It does, however, label as unscientific any hypotheses involving universes that are intrinsically, causally disjoint from ours. In any case we should avoid implying any pejorative connotations to the term "unscientific". Also, although the proper subject matter of physics may be "inanimate", one can note that the presence of an "observer" is an essential feature of quantum mechanics and that animate objects also obey the laws of physics.
Because physics is the most mathematically precise of all the sciences, physicists sometimes imply that only their methods can provide answers as to why things are as they are. In fact, it is clear that physical reasoning provides only connections between various principles and phenomena and never provides a final and fundamental answer to the question "why". Every solution in physics comes together with further questions. Consider the following typical train of physical reasoning:
(Q) Why does the moon revolve around the earth and the earth around the sun?
(A) Because of Newton's universal inverse square law of attraction.
(Q) But why is the law "inverse square"?
(A) Because the graviton is a massless particle.
(Q) But why is the graviton massless?
(A) Because all the forces of nature are gauge interactions.
(Q) But why are the forces gauge forces?
(A) Because gauge forces emerge naturally from string theory.
(Q) But why?... but why?... but why?...
Each "answer" above summarizes a possible physics seminar or even a semester-long course. Often such a train of questions can only be ended by the statement that, if things were different, we would not be here to discuss the question. This is the point at which the anthropic principle is currently entering persistently into physics discussions. By offering such an answer, perhaps we are going beyond physics into metaphysics. Perhaps this is a sign of the maturity of physics at this point in the 21st century although no one should propose that there are no further connections to be made by traditional physics methods. Perhaps we are at the point where much might be learned by expanding the above definition of physics or by defining a new and broader science. In any case, the fact that chains of questions such as the above have no final answer within physics, as currently defined, should not blind us to the fact that, in the process, we have uncovered an amazing elegance in the natural laws.
The anthropic principle, if accepted as an intrinsic property of the material universe, does provide a mode of response to questions that are beyond the current capabilities of theoretical physics or even beyond the definition of what questions physics can address. It is possible to look at each possible question about nature from the point of view of the principle. For instance, among many others, the questions of dark matter and of neutrino masses have been investigated from this point of view (7). The question of why our universe appears to have three space dimensions came under physics scrutiny with the rise of string theory but is still without any definitive, scientific answer. From the anthropic point of view, however, it has been noticed that in a universe with more than three dimensions of the nature of x, y, and z, there would be no stable planetary orbits thus precluding the evolution of life. If there were fewer than three, it has been argued that the highly connected neural networks in the human brain would not have been possible (8). Finally one can note that there are questions that have still never been asked from the scientific point of view. One of these is why nature is quantum mechanical at all rather than classical. From the anthropic point of view the answer is obvious: in classical physics the atom is unstable thus disallowing the evolution of stars, planets, and life while in quantum mechanics the atom is stable or at least much longer lived than the universe todate.
From the dawn of physics in ancient Greece up to modern times there has been a competition between two threads of thought: intelligent design vs. extreme randomness. One, championed by Anaxagoras and Aristotle holds that nothing happens without an efficient (mechanical) cause and a final cause (purpose). The ultimate cause was labeled "the Mind". The other, championed by Leucippus and Democritos, holds that everything happens by "chance" in a universe of random particle collisions. "In reality there is nothing but atoms and the void." The former point of view was dominant from the time of Aristotle until at least the advent of quantum mechanics in the 1920's. It seemed to lend itself to a theistic worldview and, indeed, it was greatly elaborated by the medieval philosophers. Quantum mechanics forced the abandonment of the notion that specific events at the atomic level have precise causes within the physical systems themselves. The laws of physics since the birth of quantum theory have been statistical in nature rather than deterministic. In reducing physical law to probability theory, it was considered natural by some to abandon notions of meaning and purpose in nature. This view seems to lend itself to a non-theistic weltanschauung but, if one insists on a causal theory, it has also a theistic interpretation involving causal agents external to the physical system.
Much of the hostility that the anthropic principle has met is due to its possible suggestion of a deep purpose built into the laws of physics and driving the evolution of the material universe. However, just as, since Galileo, believers in religion need to accommodate the truths derived from scientific experimentation, believers in the absence of an external designer might need to accommodate a fundamental principle building into the laws of physics the eventual evolution of intelligent life. Several imaginative proposals have been advanced to reconcile the anthropic principle with the assumed absence of fundamental causes or purposes.
The term "anthropic principle" was coined by the astrophysicist B. Carter (9). It can be discussed in either a "weak form" or a "strong form" which can be paraphrased as follows:
- Strong form: The elementary particles and fundamental forces are uniquely those that allow the evolution of intelligent life.
- Weak form: Among the set of all possible universes is at least one which allows the evolution of intelligent life.
We use the term universe to describe a region of space-time with a common set of physical constants, elementary particles, and fundamental forces. Both the strong form and the weak form have theistic as well as non-theistic interpretations.
The strong form claims to answer Einstein's famous question as to whether nature had a choice in setting down the laws of physics. It suggests, remarkably, that the only possible universe is the one (ours) which allows the evolution of life. It appeals to those who seek to find a unique, mathematically consistent string theory that will "explain" why things are as we see them. However, the strength of science, namely its requirement of confirmation by reproducible experiments, is also its greatest weakness namely its contingency. No matter how "unique" and compelling a theory might be, it still requires experimental confirmation which shows that no physical theory is realized necessarily in nature. If physics ends with a final string theory of everything as some hypothesize, one would still have to ask why this theory is realized rather than, say, nothing or one of the free (non-interacting) field theories which are at present the only non-perturbatively soluble mathematically consistent theories in our four dimensional space.
The weak form appeals to those who believe in ultimate randomness with no purpose or fundamental causes in nature. If some measured value, such as the cosmological constant discussed above, takes an anthropically selected value of the order of 1/n of its natural value, this might indicate the existence of at least n alternative universes which are incompatible with the evolution of life. If nature makes random quantum jumps between these alternative universes, eventually it will land in one of the few where life is possible and then life will evolve as it has on earth. If there are many "anthropic coincidences" each with an a priori probability of 1/ni, the number of alternative universes is, at least, the product of the ni. Supposedly, some string theorists are attempting to enumerate the local ground states of string theory. The cosmological constant "coincidence" suggests that there should be at least 10120 local minima in string theory. It is not clear whether this theory of a vast number of alternative universes where life is not possible is experimentally verifiable as required if it is to be a "scientific" theory in the current sense of the word. Perhaps, instead, it is part of a coming new metaphysics.
The weak form appeals to those who believe in ultimate randomness with no purpose or fundamental causes in nature. If some measured value, such as the cosmological constant discussed above, takes an anthropically selected value of the order of 1/n of its natural value, this might indicate the existence of at least n alternative universes which are incompatible with the evolution of life. If nature makes random quantum jumps between these alternative universes, eventually it will land in one of the few where life is possible and then life will evolve as it has on earth. If there are many "anthropic coincidences" each with an a priori probability of 1/ni, the number of alternative universes is, at least, the product of the ni. Supposedly, some string theorists are attempting to enumerate the local ground states of string theory. The cosmological constant "coincidence" suggests that there should be at least 10120 local minima in string theory. It is not clear whether this theory of a vast number of alternative universes where life is not possible is experimentally verifiable as required if it is to be a "scientific" theory in the current sense of the word. Perhaps, instead, it is part of a coming new metaphysics.
The rise to prominence of the anthropic principle in physics may have generated many wrong questions. The following questions (10) have been posed.
The former implicitly assumes that each question has an explanation either in terms of traditional physics methodology or in terms of an anthropic principle but not both. The second question seems to ask when one should give up trying to find a standard physics explanation and resort to the anthropic principle. It has been stated (11) that with one good standard physics explanation of the cosmological constant, for example, no one would talk about the anthropic principle in this context.
It seems to me, however, that arguments from the anthropic principle will probably always co-exist with standard physics connections to Lagrangians and other phenomena. For instance, the fact that nature must be quantum mechanical rather than classical in order for atoms and matter-based life to exist, does not detract from the value of quantum theory nor vice-versa.
Perhaps, it comes down to the question whether physics is really like the onion in the familiar analogy. Will there always be, without limit, another layer to explore after peeling off the prior layers? Alternatively, will physics eventually get to the bottom after which there are no further answers to be had apart from the anthropic principle?
- Which quantities are "anthropic"?
- How does one guard against "premature application" of the anthropic principle?
The former implicitly assumes that each question has an explanation either in terms of traditional physics methodology or in terms of an anthropic principle but not both. The second question seems to ask when one should give up trying to find a standard physics explanation and resort to the anthropic principle. It has been stated (11) that with one good standard physics explanation of the cosmological constant, for example, no one would talk about the anthropic principle in this context.
It seems to me, however, that arguments from the anthropic principle will probably always co-exist with standard physics connections to Lagrangians and other phenomena. For instance, the fact that nature must be quantum mechanical rather than classical in order for atoms and matter-based life to exist, does not detract from the value of quantum theory nor vice-versa.
Perhaps, it comes down to the question whether physics is really like the onion in the familiar analogy. Will there always be, without limit, another layer to explore after peeling off the prior layers? Alternatively, will physics eventually get to the bottom after which there are no further answers to be had apart from the anthropic principle?
(1) Case Western Reserve University Conference, The Future of Cosmology, Oct 10-12 (2003)
(2) V. Agrawal, S.M. Barr, John Donoghue, and D. Seckel, Phys. Rev. D57, 5480 (1998)
(3) S. Weinberg, Physical Review Letters 59, 2607 (1987)
(4) F. Hoyle D.N.F. Dunbar, W.A. Wenzel, and W. Whaling, Phys. Rev. 92, 1095 (1953)
(5) J. Barrow and F. Tipler, The Anthropic Cosmological Principle, Clarendon Press, Oxford, (1986)
(6) H. Oberhummer, R. Pichler, A. Csoto, arXiv:nucl-th/9810057
(7) J. Garriga and A. Vilenkin, astr-ph/0210358; M. Tegmark and A. Vilenkin, astr-ph/0304536
(8) S.M. Barr, Anthropic Coincidences, www.petersnet.net/browse/4150.htm
(9) B. Carter, IAU symposium No. 63, Confrontation of Cosmological Theories with Observational Data, edited by M. Longair (Reidel, Dordrecht)(1974)
(10) S. Dimopoulos, Conference on Supersymmetry 2003, Tucson, Arizona, June 2003
(11) D. Gross at the Case Western Reserve University Conference, The Future of Cosmology, Oct 10-12 (2003)
L. Clavelli, November 2003
(2) V. Agrawal, S.M. Barr, John Donoghue, and D. Seckel, Phys. Rev. D57, 5480 (1998)
(3) S. Weinberg, Physical Review Letters 59, 2607 (1987)
(4) F. Hoyle D.N.F. Dunbar, W.A. Wenzel, and W. Whaling, Phys. Rev. 92, 1095 (1953)
(5) J. Barrow and F. Tipler, The Anthropic Cosmological Principle, Clarendon Press, Oxford, (1986)
(6) H. Oberhummer, R. Pichler, A. Csoto, arXiv:nucl-th/9810057
(7) J. Garriga and A. Vilenkin, astr-ph/0210358; M. Tegmark and A. Vilenkin, astr-ph/0304536
(8) S.M. Barr, Anthropic Coincidences, www.petersnet.net/browse/4150.htm
(9) B. Carter, IAU symposium No. 63, Confrontation of Cosmological Theories with Observational Data, edited by M. Longair (Reidel, Dordrecht)(1974)
(10) S. Dimopoulos, Conference on Supersymmetry 2003, Tucson, Arizona, June 2003
(11) D. Gross at the Case Western Reserve University Conference, The Future of Cosmology, Oct 10-12 (2003)
L. Clavelli, November 2003
