An exchange of emails with Roberto Mangabeira Unger concerning temporality, the reality of the future, and how such things relate to physics (including some examples from causal set theory). 1. Sorkin to Unger, May 2010 Subject: Question on your notion of time Date: Sat, 29 May 2010 18:15:51 -0400 From: Rafael D Sorkin To: Roberto Unger Dear Roberto, Although I was not part of last week's conference here on "laws of nature", I did view your remarks online, and they left me confused. If you have an opportunity, I'd appreciate your clearing up an apparent contradiction in what you said. On one hand you stated fairly emphatically that the "block universe" concept was an enemy to be combatted. On the other hand you also said at some point that "The future is real". To the extent that I can attach a definite meaning to these two statements, they seem to me to be direct opposites. Did you perhaps mean to say "The future *will be* real" rather than "The future *is* real"? Or were you posing some sort of dialectical contradiction for subsequent thought to overcome? Or have I perhaps just missed some more obvious interpretation of your statements? In any case, I'm hoping you can clear this up, since much of your presentation resonated with the type of thinking I've been led to by my own work on the problem of quantum gravity, and more specifically on the causal set program. Cordially, Rafael Sorkin 2. Unger to Sorkin, May 2010 Subject: CORRECTED VERSION OF THE RESPONSE I SENT YOU YESTERDAY: THE REALITY OF TIME AND THE REALITY OF THE FUTURE Date: Tue, 1 Jun 2010 10:57:28 -0400 From: Roberto Mangabeira Unger May 31, 2010 Dr. Rafael D. Sorkin Perimeter Institute for Theoretical Physics Dr. Sorkin, Thank you for your letter. I doubt that we disagree about the matters you address, if I understand correctly both your question and your perspective. Your query illustrates the endless confusion to which references to "being" or "reality" give rise. The confusions brought out by mistaking predication for a claim of existence are familiar; the confusions generated by the temporal setting of claims of existence remain poorly understood. The most important reason for this situation is that there lurks under the second set of confusions an unresolved struggle over the fact of the matter about the reality of time. 1. We live only in the present; all of our experience is a sequence of present moments. To say as much is to acknowledge that consciousness entraps us in a highly restrictive experience of time. We nevertheless have the power to attenuate some of the implications of this entrapment. One of the many ways in which we can do so is to correct for the tyranny of the present moment over our consciousness. We can -- and should -- recognize that the world is not limited in the way we are. 2. Everything that is or will ever be true in the world must be true in some moment. 3. When we allow the illusions of consciousness to be corrected by the work of self-reflection and by the discoveries of science, we have reason to conclude that nothing is more real than time, stretching into the past and into the future. 4. What is indisputably true is that the world, such as we experience it in the present moment, has both a past and a future. That it has a future is just as real as that it has a past. This is all I meant by saying that the future is real. What is excluded by this simple idea is philosophical "presentism," which would reduce the structure of the world, and therefore as well the nature of time, to the structure of consciousness, with its characteristic inability to escape entrapment in the present moment. 5. There is, however, a stark asymmetry between past and future. Past events cannot be undone; that is the basis of all tragedy in our experience; they are defined and gone. But the content of the future is open in both a subjective and an objective sense. It is open in a subjective sense because we encounter intractable difficulties in knowing it. But it is also open in a objective sense if we admit, as I believe we must, that everything, including the laws of nature and the most fundamental structural features of reality, changes sooner or later. However, the precise sense in which the future is open is not something that can be inferred from philosophical ideas; it depends on what, through science, we find out about the world. Such a view is incompatible with the "block universe" picture or, more generally, with any attempt to extrapolate the "Newtonian paradigm," with its characteristic reliance on initial conditions and on law governed phenomena within a configuration space bounded by those conditions, into an approach to the world as a whole. Laplacean determinism is simply the extreme form of the ideas implicit in the Newtonian paradigm and in the block universe view. It prompts us to imagine ourselves looking at the world from the outside, from the heights of supposedly timeless laws of nature. And it inspires a conception of knowledge according to which if only we understood the world well enough (as God understands it) we would see future events already contained in past ones, with the result that time would lose much of its reality. We would see, as God is imagined to see, only the eternal now. It does not, however, follow from the rejection of such notions that we are entitled to say any of the things that the feel-good metaphysics would like to say: that the world is open always, or increasingly, to our designs. 6. The usage you recommend -- "the future will be real" in lieu of "the future is real" -- to avoid any inference from the reality of the future to its closure, or to the predetermination of future events, has advantages and disadvantages. Its advantage is to avoid the misinterpretation that I have inadvertently provoked. Its disadvantage is to fail to do justice to the reality of the temporal continuum -- of global time, going all the way down and holding sway over everything -- as an objective feature of the world, even as its most important feature. So an alternative approach is simply to say that past, present, and future are all real, but then to be more explicit than I was in that talk and discussion in fixing the meaning of the words. 7. Time is not like a film, made up of still photographs. It is not an accumulation of slices. Insistence on speaking as if it were like a film arises from the widespread antipathy to an unconditional recognition of the reality of time. However, the deep problem revealed in the film image of time is the difficulty that all our non-causal categorical schemes -- beginning with our logical and mathematical reasoning -- have in dealing with the temporal continuum. The slice or film language is yet another way to subordinate time to the anti-temporal biases of those forms of reasoning. It is remarkable that those who have acknowledged the intimate relation between time and causation, in a line that goes from Leibniz to Hans Reichenbach and beyond, have nevertheless had trouble fully conceding to time the character of a continuum. They retreat, despite themselves, to the conception of a series of still photographs. The dialectic of their halting advances and of their involuntary retreats is such that the idea of the continuum becomes, for reasons different from the ones that Leibniz adduced, a labyrinth. 8. I suspect that these contradictory impulses help explain the bewilderment that may be aroused by an attempt to dissociate talk of the future as real from the block universe view or from the Laplacean ideal of comprehensive, time-eviscerating determinism. What is real is not exhausted by what exists now. The world was once different in its structure and workings from what it is like now, and we can expect it to be different again in the future. The motto -- were it not for the unspeakable disproportion between human and cosmic time -- should be: you haven't seen anything yet. We have no difficulty in saying that something is real although it is distant from us in space. However, the assertion that it can be real although distant from us (who live only in the present) in time may cause puzzlement to contemporary scientists. If you asked people on the street whether the future is real (not whether it will be real), they would, I believe, overwhelmingly say, startled by the question: of course, it is real. I doubt whether one in ten would associate the reality of the future with the total predetermination of future events by past events. The reason why this common usage provokes misunderstanding among physicists may have something to do with a tradition of thought resistant to the full recognition of the reality of time -- a tradition powerful enough to influence its enemies as well as its votaries. Once these misunderstandings are cleared up, and any disagreement associated with them is either dispelled or rendered precise, we are free to turn to what matters: the immense substantive enigmas posed by time. 1. A major project of twentieth-century physics, tying it to much of what preceded it, has been the spatialization of time, The influence of this project is many-sided and omnipresent; general relativity is only its most consummate expression. It views time as emergent or even, at the geometry-worshipping extreme of its explanatory program, as illusory. 2. Yet this same twentieth-century physics laid, in cosmology, the basis for the idea of a history of the universe.There now seems to be more reason to believe in a succession of causally connected worlds than in a plurality of worlds causally unconnected to one another. Properly understood, the idea of a history of the universe undermines the project of the spatialization of time. 3. The implications of universal history appear to be better accommodated within a view that affirms a global time in a single world, admitting of succession but not of plurality. According to such a view, everything that has ever happened, or that will ever happen, in the world can be placed on a single time chart, notwithstanding the objections to simultaneity arising out of special relativity. Time is not emergent, although space may well be, with the result that we would more justified in temporalizing space than in spatializing time. The idea of a timeless framework of laws of nature must be overthrown, as the idea of a space-time background, separable from the phenomena, was overthrown in the twentieth century. The transformation of transformation -- the ways in which phenomena change themselves change -- should come to be seen as a defining attribute of the radical and inclusive reality of time. A historical account must be viewed as more fundamental than a structural analysis rather than as derivative from it. And an agenda of observation and experiment must be devised that translates these conceptions into a workable agenda of empirical inquiry, with the development of observational and experimental equipment redirected accordingly. 4. Now it is true that any such view immediately confronts the countervailing fact that the laws of nature appear to have been stable from early in the history of the present universe. However, this fact, if it is a fact, can be given a very different reading. Let me state this alternative account only metaphorically, in pre-scientific and pre-philosophical language. We may imagine the present, cooled-down universe, in comparison to the universe in its fiery and formative stage, as a kind of living corpse: with relatively limited kinetic energy, temperature, and degrees of freedom, with a relatively consolidated structure, and with stable regularities -- the laws of nature. Yet there was a time when the distinction between states of affairs and regularities was unclear (a circumstance that can be described alternately as a time of law-giving or of lawlessness), when the present structural division of the world into well-defined constituents was not yet established, and when the phenomena were excited to much higher degrees of freedom than those enjoyed by the living corpse. The unexplained values of the dimensionless constants or parameters of nature may have their origin in the process from which this formative moment gave way to the ensuing regularities and structures. 5. Any such view presents us with a problem that Smolin and I have provisionally called the meta-law conundrum. If the change of the laws is itself law-governed, we have reintroduced the idea of a timeless framework. If the change of the laws is not law-governed, not even in the sense of statistical determination, it may seem to be uncaused or arbitrary. A possible solution to the conundrum of the meta-laws -- and one that has many counterparts in the life sciences and in social and historical study -- is that the laws develop, or change, coevally with the phemomena. Under such an account, causal connection is stressed but not broken. 6. To an outsider, however, there seems to be no adequate foundation in the present structure of physics and cosmology for such a view. The exorbitant claims made on behalf of entropy dynamics as a basis for a universal view of time appear to be confused. Thermodynamics and hydrodynamics are regional theories: they address parts or regions of nature. They foresee, in those parts or regions, processes that are in principle reversible and that become irreversible only in the context of particular initial conditions and therefore of a particular history: not just the history resulting from the initial conditions but also the history resulting in the initial conditions. These theories apply only through the familiar Newtonian practice of defining initial conditions and a configuration space of law-governed phenomena -- a practice that cannot be legitimately generalized to the world as a whole. On the other hand, the attempt to ground a view of global, irreversible and non-emergent time on the quantum-mechanical description of the smallest present constituents of nature seems misguided for a different reason. The structural outcome of a historical process can provide fragmentary clues to the understanding of that process but it cannot serve as the basis for a general theory of the transformation that produced the outcome. In principle, it seems wrong-headed to expect that such a theory could ever arise as the next stage in the now prevailing scientific agenda of unification of the laws of nature -- an agenda thoroughly wedded to the spatialization of time and to the dismissal of history as secondary to structure. A revolutionary reorientation is required. Not only would such a reorientation reshape the established program of cosmology and physics; it would also change our understanding of mathematics and of its relation to nature and to science. Immanuel Kant was surely right in saying that when a disagreement about the meaning of words has gone on for a long time in the history of thought, it will always be found that behind it there is a real dispute about the world. I hope that the problems and clarifications mentioned in the first part of this answer to your message will not prevent us from devoting ourselves to the fundamental issues addressed in the second part. And I thank you once again for your attention and engagement. Roberto Mangabeira Unger 3. Sorkin to Unger, Jun 2010 Subject: Re: RESPONSE TO YOUR LETTER: THE REALITY OF TIME AND THE REALITY OF THE FUTURE Date: Thu, 3 Jun 2010 01:18:52 -0400 In-Reply-To: your message of Mon, 31 May 2010 10:31:37 -0400 From: "Rafael D Sorkin" To: Roberto Mangabeira Unger Dear Roberto, Many thanks for your very helpful and suggestive exposition. We do indeed agree on many basic points, but the more I reread what you have written, the more I feel that there remain important divergences as well. How deep the differences go is hard to tell, because their real nature would only reveal itself in the different forms of future scientific theory they would correspond to. Rather than attempt to systematize my own thinking in the way you have done (which would probably delay my answer indefinitely) I will just enumerate some points on which I think we might differ, and then interpolate several further remarks in the text of your message. I can also refer you to an article of mine that bears on some of these questions. It is entitled "Relativity-theory does not imply that the future already exists: a counterexample" and its most up to date version can be found at either of the following (tenth paper from the bottom): http://www.perimeterinstitute.ca/personal/rsorkin/some.papers/ http://www.physics.syr.edu/~sorkin/some.papers/ On the same web page, the paper called "Indications of causal set cosmology" perhaps offers an example of "transformation of transformation", though probably not as radical a one as you'd like. 1. For my part, I conceive of "existence" as an ongoing process of growth -- or in a more definite mathematical framework, of *births* of causal set elements. This process is inherently (and crucially for the recovery of spacetime) "asynchronous" in the sense that two births can be "spacelike" to each other, i.e. without direct causal/temporal connection. In such a case neither element is the causal ancestor of the other. Is this notion of an asynchronous and discrete birth-process compatible with your idea of time? Some of your remarks suggest not, especially what you say about temporal "moments" and about time being a continuum. 2. Although the idea of reality as an -- unfinished -- birth process does seem to me to clash with the "block universe", it's not clear that the same can be said of indeterminism or the openness of the future. That is, I don't see why the block universe idea cannot live happily with the idea of an open future. The latter could mean merely that the dynamical laws failed to determine the future, even given the past in all its detail. Theories of stochastic processes such as Brownian motion are precisely like that, without in any way contradicting the "spatialization of time" (unless perhaps probability itself contradicts the spatialization of time). 3. Although I appreciate that this is not at all where you are trying to go, I feel that in some places you come close to implying that it is purely our consciousness that brings temporality to the world, much as advocated by (I think) Gruenbaum in trying to reconcile "becoming" with the block universe. In contrast I would prefer to say that (the passage of) time is the continuous birth of elements of the causal set. Granted I cannot make the connection concretely, but I like to think that our "experience of the now" is indirectly reflecting this profusion of ongoing microscopic events that (or whose "record") constitute(s) reality. 4. You talk a lot about "laws" but little about "substance" ("ontology", "kinematics"). To me the latter is even more basic than the former. If the one has historical aspects, then so must the other. I suspect you feel the same way, even if you haven't emphasized it. These seem to me to be the most salient divergences, or potential divergences, between us. The following interpolations might help flesh out some of them and also bring in some lesser ones. > Dr. Sorkin, > Thank you for your letter. I doubt that we disagree about the > matters you address, if I understand correctly both your question and your > perspective. > Your query illustrates the endless confusion to which references to > "being" or "reality" give rise. The confusions brought out by mistaking > predication for a claim of existence are familiar; the confusions generated > by the temporal setting of claims of existence remain poorly understood. > The most important reason for this situation is that there lurks under the > second set of confusions an unresolved struggle over the fact of the > matter about the reality of time. > 1. We live only in the present; all of our experience is a sequence of > present moments. To say as much is to acknowledge that > consciousness entraps us in a highly restrictive experience of time. We > nevertheless have the power to attenuate some of the implications of this > entrapment. One of the many ways in which we can do so is to correct for > the tyranny of the present moment over our consciousness. We can -- and > should -- recognize that the world is not limited in the way we are. > 2. Everything that is or will ever be true in the world must be true > in some moment. Does this last sentence conflict with "asynchronous becoming"? (See my point 1 above? > 3. When we allow the illusions of consciousness to be corrected by the > work of self-reflection and by the discoveries of science, we have reason > to conclude that nothing is more real than time, stretching into the past > and into the future. > 4. What is indisputably true is that the world, such as we experience > it in the present moment, has both a past and a future. That it has a > future is just as real as that it has a past. This is all I meant by saying > that the future is real. What is excluded by this simple idea is > philosophical "presentism," which would reduce the structure of the world, > and therefore as well the nature of time, to the structure of > consciousness, with its characteristic inability to escape entrapment in > the present moment. I heartily agree with you about "presentism" and entrapment in the present moment. Indeed, I'm not sure "the present" has any being at all, other than as the process by which the past is built up. Nevertheless I do think there is something special about the (local) present, and to that extent I believe that our forms of experience or "consciousness" are not misleading us. (Compare the Vibhajavadin school of Buddhist philosophers, who are said to have held that what's real is the present together with that part of the past that has not yet lost its causal efficacy.) > 5. There is, however, a stark asymmetry between past and future. Past > events cannot be undone; that is the basis of all tragedy in our > experience; they are defined and gone. But the content of the future is > open in both a subjective and an objective sense. > It is open in a subjective sense because we encounter intractable > difficulties in knowing it. But it is also open in a objective sense if we > admit, as I believe we must, that everything, including the laws of nature > and the most fundamental structural features of reality, changes sooner or > later. However, the precise sense in which the future is open is not > something that can be inferred from philosophical ideas; it depends on > what, through science, we find out about the world. > Such a view is incompatible with the "block universe" picture or, But are they really incompatible? See my comment 2 above. More generally, is there no more to the distinction between future and past than the "openness" of the one vs. the "closedness" of the other? > more generally, with any attempt to extrapolate the "Newtonian paradigm," > with its characteristic reliance on initial conditions and on law governed > phenomena within a configuration space bounded by those conditions, into an > approach to the world as a whole. Laplacean determinism is simply the > extreme form of the ideas implicit in the Newtonian paradigm and in the > block universe view. It prompts us to imagine ourselves looking at the > world from the outside, from the heights of supposedly timeless laws of > nature. And it inspires a conception of knowledge according to which if > only we understood the world well enough (as God understands it) we would > see future events already contained in past ones, with the result that time > would lose much of its reality. We would see, as God is imagined to see, > only the eternal now. Yes. And it's this "God's view" (together, possibly, with the assumption of a continuum) that misleads people into thinking that general covariance forces the block universe on us. > It does not, however, follow from the rejection of such notions > that we are entitled to say any of the things that the feel-good > metaphysics would like to say: that the world is open always, or > increasingly, to our designs. > 6. The usage you recommend -- "the future will be real" in lieu of > "the future is real" -- to avoid any inference from the reality of the > future to its closure, or to the predetermination of future events, has I think I mean more than just indetermination of the future... > advantages and disadvantages. Its advantage is to avoid the > misinterpretation that I have inadvertently provoked. Its disadvantage is > to fail to do justice to the reality of the temporal continuum -- of global > time, going all the way down and holding sway over everything -- as an > objective feature of the world, even as its most important feature. So an It's this "disadvantage" that I fail to grasp. Why does the reality of time militate for or against the reality of the future? > alternative approach is simply to say that past, present, and future are > all real, but then to be more explicit than I was in that talk and > discussion in fixing the meaning of the words. > 7. Time is not like a film, made up of still photographs. It is > not an accumulation of slices. I couldn't agree more. Such a view is not even compatible with relativity theory. > not an accumulation of slices. Insistence on speaking as if it were like a > film arises from the widespread antipathy to an unconditional recognition > of the reality of time. However, the deep problem revealed in the film > image of time is the difficulty that all our non-causal categorical schemes > -- beginning with our logical and mathematical reasoning -- You are quite right that present-day physics knows no fundamentally causal categories. This, incidentally, impedes the task of formulating relativistic causality "intrinsically", without reference to human (or similar) agents. > -- beginning with our logical and mathematical reasoning -- have in dealing > with the temporal continuum. The slice or film language is yet another way > to subordinate time to the anti-temporal biases of those forms of > reasoning. > It is remarkable that those who have acknowledged the intimate > relation between time and causation, in a line that goes from Leibniz to > Hans Reichenbach and beyond, have nevertheless had trouble fully conceding > to time the character of a continuum. Your use of the word continuum bothers me because I work on the hypothesis that the deep structure of spacetime is discrete. Are you trying to say that moments of time are not "separable" from each other? But I believe that the interlocking pattern of causal relationships among "atoms of spacetime" (in a causal set) can "bind" them together just as effectively as continuity can. > to time the character of a continuum. They retreat, despite themselves, to > the conception of a series of still photographs. The dialectic of their > halting advances and of their involuntary retreats is such that the idea > of the continuum becomes, for reasons different from the ones that Leibniz > adduced, a labyrinth. > 8. I suspect that these contradictory impulses help explain the > bewilderment that may be aroused by an attempt to dissociate talk of the > future as real from the block universe view or from the Laplacean ideal of > comprehensive, time-eviscerating determinism. What is real is not exhausted > by what exists now. The world was once different in its structure and > workings from what it is like now, and we can expect it to be different > again in the future. The motto -- were it not for the unspeakable > disproportion between human and cosmic time -- should be: you haven't seen > anything yet. > We have no difficulty in saying that something is real although it > is distant from us in space. However, the assertion that it can be real > although distant from us (who live only in the present) in time may cause > puzzlement to contemporary scientists. But what is this, if not the spatialization of time? > puzzlement to contemporary scientists. If you asked people on the street > whether the future is real (not whether it will be real), they would, I > believe, overwhelmingly say, startled by the question: of course, it is > real. I once did ask people about this! Most were indeed startled by the question, but they usually said that neither past nor future, but only the present was real. My sample was pretty homogeneous culturally, though. > real. I doubt whether one in ten would associate the reality of the future > with the total predetermination of future events by past events. The reason > why this common usage provokes misunderstanding among physicists may have > something to do with a tradition of thought resistant to the full > recognition of the reality of time -- a tradition powerful enough to > influence its enemies as well as its votaries. > > Once these misunderstandings are cleared up, and any disagreement > associated with them is either dispelled or rendered precise, we are free > to turn to what matters: the immense substantive enigmas posed by time. > 1. A major project of twentieth-century physics, tying it to much of > what preceded it, has been the spatialization of time, The influence of > this project is many-sided and omnipresent; general relativity is only its > most consummate expression. It views time as emergent or even, at the > geometry-worshipping extreme of its explanatory program, as illusory. Not everyone has drawn such conclusions from general relativity. For example Howard Stein has not. To some extent the formalism does point in this direction, but to some extent I think people have just hung their metaphysical biases on it. > 2. Yet this same twentieth-century physics laid, in cosmology, the > basis for the idea of a history of the universe. Good point! > basis for the idea of a history of the universe.There now seems to be more > reason to believe in a succession of causally connected worlds than in a > plurality of worlds causally unconnected to one another. Properly > understood, the idea of a history of the universe undermines the project of > the spatialization of time. > 3. The implications of universal history appear to be better > accommodated within a view that affirms a global time in a single world, > admitting of succession but not of plurality. According to such a view, > everything that has ever happened, or that will ever happen, in the world > can be placed on a single time chart, notwithstanding the objections to > simultaneity arising out of special relativity. I agree, but only if you allow me to add (precisely because of the objections you cite) that such a "time chart" cannot be a "time line": temporality is not one-dimensional. > simultaneity arising out of special relativity. Time is not emergent, > although space may well be, with the result that we would more justified in > temporalizing space than in spatializing time. Agreed. > temporalizing space than in spatializing time. The idea of a timeless > framework of laws of nature must be overthrown, as the idea of a space-time > background, separable from the phenomena, was overthrown in the twentieth > century. The transformation of transformation -- the ways in which > phenomena change themselves change -- should come to be seen as a defining > attribute of the radical and inclusive reality of time. A historical > account must be viewed as more fundamental than a structural analysis > rather than as derivative from it. And an agenda of observation and > experiment must be devised that translates these conceptions into a > workable agenda of empirical inquiry, with the development of observational > and experimental equipment redirected accordingly. I think there's a more general problem with laws of nature that you only hint at here with your phrase "separable from the phenomena". Namely physical laws (at least in the way we currently formulate them) appear to "govern events from outside". Such a framework of external laws seems fundamentally wrong to me, but where is the alternative? In the case of human laws, we understand -- mas o menos -- how they arise and how they are enforced. In the case of natural laws we apparently lack any such understanding. Without it, I see no way to make real progress on things like "the transformation of transformation". With it, on the other hand, I'm not sure that your opposition between historical accounts and structural analysis will survive. Therefore I have no idea how to proceed further. From where will this new "agenda of empirical inquiry" arrive? With what new instruments of observation and experiment? This reminds me of the feminist critique of science. On one hand, one feels some justice in the criticism. On the other hand, what would a less chauvinist science look like? Should we for example relinquish the demand that experiments and observations be repeatable? Then what? If it's a revolution we need, perhaps we must wait for the contradictions to ripen. Perhaps we are still in the early stages of utopian socialism, as it were. Maybe all we can do for now is to keep the critique alive, without being able to realize the changes it is calling for. On the other hand, it seems possible to take small steps in the direction indicated, as in the paper of mine on cosmology that I cited above. Similarly, there is some prospect at least to unify "laws of motion" with "initial conditions", via the path integral, even if that doesn't yet unify dynamics with kinematics... > 4. Now it is true that any such view immediately confronts the > countervailing fact that the laws of nature appear to have been stable from > early in the history of the present universe. However, this fact, if it is > a fact, can be given a very different reading. I think it is a fact. And I agree with you that it can, and arguably must be read differently. If, for example, the cosmos began as a very small thing with minimal structure, then how could it possibly have known how to implement the very elaborate laws of, for example, the standard model? The "material basis" for such laws was just not present at that early stage. > a fact, can be given a very different reading. Let me state this > alternative account only metaphorically, in pre-scientific and > pre-philosophical language. We may imagine the present, cooled-down > universe, in comparison to the universe in its fiery and formative stage, > as a kind of living corpse: with relatively limited kinetic energy, > temperature, and degrees of freedom, with a relatively consolidated > structure, and with stable regularities -- the laws of nature. Yet there > was a time when the distinction between states of affairs and regularities > was unclear (a circumstance that can be described alternately as a time > of law-giving or of lawlessness), when the present structural division of > the world into well-defined constituents was not yet established, and when > the phenomena were excited to much higher degrees of freedom than those > enjoyed by the living corpse. The unexplained values of the dimensionless > constants or parameters of nature may have their origin in the process from > which this formative moment gave way to the ensuing regularities and > structures. > 5. Any such view presents us with a problem that Smolin and I have > provisionally called the meta-law conundrum. If the change of the laws is > itself law-governed, we have reintroduced the idea of a timeless framework. > If the change of the laws is not law-governed, not even in the sense of > statistical determination, it may seem to be uncaused or arbitrary. A > possible solution to the conundrum of the meta-laws -- and one that has > many counterparts in the life sciences and in social and historical study > -- is that the laws develop, or change, coevally with the phemomena. Even more so, the laws should be *integrated* with the phenomena (in the sense of the material occurrences). They should not stand outside matter. If this were possible, their evolution would follow almost automatically. > -- is that the laws develop, or change, coevally with the phemomena. Under > such an account, causal connection is stressed but not broken. > 6. To an outsider, however, there seems to be no adequate foundation > in the present structure of physics and cosmology for such a view. I'm afraid you are right. We have no way to formulate laws that are fully conditioned by history, probably for reasons I have alluded to above. On the other hand, we do have examples of "emergent" laws -- together with emergent forms or states of matter -- that arise from historical processes. Think for example of phase transitions and hysteresis. > in the present structure of physics and cosmology for such a view. The > exorbitant claims made on behalf of entropy dynamics as a basis for a > universal view of time appear to be confused. Thermodynamics and > hydrodynamics are regional theories: they address parts or regions of > nature. They foresee, in those parts or regions, processes that are in > principle reversible and that become irreversible only in the context of > particular initial conditions and therefore of a particular history: not > just the history resulting from the initial conditions but also the history > resulting in the initial conditions. These theories apply only through the > familiar Newtonian practice of defining initial conditions and a > configuration space of law-governed phenomena -- a practice that cannot be > legitimately generalized to the world as a whole. > On the other hand, the attempt to ground a view of global, > irreversible and non-emergent time on the quantum-mechanical description of > the smallest present constituents of nature seems misguided for a different > reason. The structural outcome of a historical process can provide > fragmentary clues to the understanding of that process but it cannot serve > as the basis for a general theory of the transformation that produced the > outcome. Why so pessimistic on this point? We do know something about the early cosmos, and we have the contradiction between general relativity and quantum mechanics to impel us forward. That could let us reach at least a partial, relative understanding of "the transformation that produced the outcome" in this case. > In principle, it seems wrong-headed to expect that such a theory > could ever arise as the next stage in the now prevailing scientific agenda > of unification of the laws of nature -- an agenda thoroughly wedded to the > spatialization of time and to the dismissal of history as secondary to > structure. A revolutionary reorientation is required. Not only would such a > reorientation reshape the established program of cosmology and physics; it > would also change our understanding of mathematics and of its relation to > nature and to science. Can you think of any example of how such a reorientation would go? Perhaps a little piece of a new kind of mathematics or physics, even if only very lightly sketched out? > > Immanuel Kant was surely right in saying that when a disagreement > about the meaning of words has gone on for a long time in the history of > thought, it will always be found that behind it there is a real dispute > about the world. Very true! Closer to the mark than the opposite claim one often hears that metaphysical disputes are about words only. > about the world. I hope that the problems and clarifications mentioned in > the first part of this answer to your message will not prevent us from > devoting ourselves to the fundamental issues addressed in the second part. > And I thank you once again for your attention and engagement. > Roberto Mangabeira Unger Let me mention some readings that might interest you if you are not already familiar with them. First there's Buddhist metaphysics, whose foundational conception of "conditioned origination" comes as close as anything I've ever encountered to describing existence as a succession of causally related births. (Incidentally, according to Stcherbatsky, Buddhist logicians pointed out a reason why mathematics could not do justice to reality, namely because it is essentially static.) Milic Capek stressed what I've called the asynchronous nature of temporal reality, which relativity theory has taught us about. Someone who said perceptive things about the historical and dialectical character of science (though perhaps not of nature itself?) was Mituo Taketani. Finally, you might enjoy (if only as a "straw man") an essay by Poincare' claiming that science has no chance of accepting the view that laws can change. Regards, Rafael P.S. If what's quoted above doesn't quite agree with your corrected message, it's because I had begun to prepare my reply before your corrections arrived and had to transfer them by hand into my draft. 4. Unger to Sorkin, May 2012 From: Roberto Mangabeira Unger To: "rsorkin@perimeterinstitute.ca" Subject: PASSAGE ATTACHED Date: Mon, 21 May 2012 18:55:35 +0000 Dr. Raphael Sorkin Raphael, You may remember that we exchanged ideas about time. I take the liberty of sending you a brief passage from the first, introductory chapter of a work in natural philosophy that I have been developing, for some time, with Lee Smolin. It is now, I hope, near completion. The unconventional method by which we proceed is that each of us writes a complete version of the entire argument. The two versions are to be published within the covers of the same book. The attached section is one about which I would especially value your comments. It may also be of particular interest to you. Best regards, Roberto 5. Sorkin to Unger, June 2012 From: Rafael D Sorkin To: Roberto Mangabeira Unger Subject: Re: PASSAGE ATTACHED X-Mailer: MH-E 8.2; nmh 1.2; GNU Emacs 23.1.1 In-reply-to: your message of Mon, 21 May 2012 18:55:35 +0000 Date: Fri, 01 Jun 2012 00:38:03 -0400 Dear Roberto, I've just finished reading your draft, and as before I feel a strong affinity for your viewpoint, in particular concerning what you call the "singularity of the world" and the "reality of time". About the limitations of mathematics, I'm less in agreement, first because I'm not convinced that mathematics inherently rejects temporality/Werden, and second because of its unquestioned success in physics so far. (What would we put in its place?) Nor would I apply the word "counterfeit" to the physicist's conception of reality. On the other hand, the idea that present-day "laws of motion" are emergent and had no foothold in the early universe seems absolutely correct to me, and I also share your preference for a "succession of universes" over a "plurality" of them. At a less consequential level, I would question your claim that causal relationships are commonly thought of as mental constructs. Whatever philosophers might say, I believe that physical scientists think of "causal links" as "real connections in nature", albeit you are correct that they mostly speak in terms of laws rather than causality. This also raises the question whether your notion of "law" isn't more restrictive than the one employed in physics. It's true that some idea of "recurrent phenomena" (manifested above all in the persistence of spacetime) does underly our notion of law, but our most general laws also make sense even if no situation ever repeats itself. In particular this holds for the classical field theories of electromagnetism and gravity. It even holds for Newtonian gravity. When you write that there may be causality without laws, I'm unsure what you mean. I'm also unsure whether I can explain why I'm unsure, but let me try. On one hand there can certainly be "temporality" without laws, if one means something like a Lorentzian manifold not obeying any particular field equations, or a very young and therefore very small causal set. One can justifiably describe this kind of temporality as "causal", as in the phrase "causal set". On the other hand, some people might reserve the word causal for the "laws of motion" that putatively govern the development of "temporality" in this sense, eg the Einstein equations in the case of spacetime. In other words, I think that there's an ambiguity in your distinction between causality and laws that might need to be resolved. Alternatively, it might need to be brought out more clearly and emphasized, if it is too inherent in the subject to be resolved. This ambiguity clouds, for me, your concept of "causality without laws". Since the latter is a central component of your argument, it would be nice to have a concrete example of it "in action", if it's not already in the book somewhere. Another notion I have trouble with is that of the "adjacent possible", also the companion concept of "state of affairs". As you describe these notions, they seem to me to partake too much of linear time, as opposed to the linear-cum-parallel time of relativity and causal sets. By the way, your comments on Leibniz also apply very well to Mach, who has more directly influenced the thinking of present-day physics. I'm unsure whether you received my previous long email of a few months ago, but in it, I expressed a doubt that I still feel: that some of the conceptions which you (and I too) are striving toward remain still too unformed to be a useful guide for physics. Quoting from that email: If it's a revolution we need, perhaps we must wait for the contradictions to ripen. Perhaps we are still in the early stages of utopian socialism, as it were. Maybe all we can do for now is to keep the critique alive, without being able to realize the changes it is calling for. On the other hand, it seems possible to take small steps in the direction indicated, as in the paper of mine on cosmology that I cited above. Similarly, there is some prospect at least to unify "laws of motion" with "initial conditions", via the path integral, even if that doesn't yet unify dynamics with kinematics... I hope these comments are helpful. Regards, Rafael 6. Unger to Sorkin, June 2012 From: Roberto Mangabeira Unger To: Rafael D Sorkin Subject: RESPONSE TO YOUR LETTER ABOUT CAUSALITY WITHOUT LAWS. DRAFT ON MATHEMATICS ATTACHED Date: Fri, 1 Jun 2012 16:53:56 +0000 June 1, 2012 Dr. Rafael D. Sorkin Dear Rafael Thanks so much for your letter. Here, by way of retaking our exchange, is a preliminary response. 1. I do not attack mathematics nor do I believe that physics can dispense with it. I do propose a conception of mathematics and of its relation to nature and to science. Three connected sets of questions arise. a. What is the best view of mathematics and of its relation to science and to nature? b. How can we account for the effectiveness of mathematics in science without surrendering to the idea that mathematics offers a shortcut to the secrets of nature (vulgar Platonism)? c. What understanding of mathematics accords with the theses of the singular existence of the universe and of the inclusive reality of time? I attach a draft of my chapter on mathematics. 2. I do not see how one can dispute that the conceptual relations among moves in logical and mathematical reasoning are atemporal, although their thinking out is always a mental event in time. A conclusion is simultaneous with its premise. The same holds for the relations among mathematical propositions, even when (as famously with the calculus) they are used to represent events in time. Causation, however, is time-bound. A cause precedes its effect. This simple contrast has radical and misunderstood implications. Reckoning with these implications must be one of the starting points for any effort to deal with the role of mathematics in science. 3. That scientists see causal links as real connections in nature seems to me to describe only half of a confusion. Scientists may tend to view causal connections as real, rather than as mental constructs in the fashion of Hume and Kant. However, they are also accustomed to regard causal connections as instances of the regular, law-like operations of nature. The statement of such laws in the form of a series of equations has been the commanding project of physics. The history of physics is, in large part, the career of those equations. However, the history of the universe being what it is, causal connections can exist without recurring over a differentiated structure of reality. They can therefore fail to assume repetitious, law-like form. Such a way of thinking that cannot be reconciled with the conventional -- and confused -- view of causation that prevails within and outside science. The operational ideology of science seems to be that both the causal connections and the laws of nature, which they supposedly enact, are real. I argue that the causal connections are more real than the laws. Their reality, as primitive features of nature, is related to the reality of time. I would like to understand the bearing of your ideas about causal sets on these issues. 4. I disagree that the argument of which these ideas form part is unformed or premature. There are, however, two problems. a. As always, with the introduction of a new set of ideas, there is no available and reliable language in which adequately to express them. It seems unwise either to rely unreservedly on the existing language or to invent a wholly new vocabulary. One needs both to use and to revise the available language, in context and cumulatively, giving new meanings to old words. b. It is vital to develop the agenda of empirical and experimental research to which the ideas lead. Otherwise, they will remain impotent natural philosophy. Best regards, Roberto 7. Sorkin to Unger, June 2012 From: Rafael D Sorkin To: Roberto Mangabeira Unger Subject: Re: RESPONSE TO YOUR LETTER ABOUT CAUSALITY WITHOUT LAWS. In-reply-to: your message of Fri, 1 Jun 2012 16:53:56 +0000 Date: Sun, 03 Jun 2012 00:58:37 -0400 Dear Roberto, Thanks for your email. Below are some extracts from it, followed by my responses to them. > I would like to understand the bearing of your ideas about > causal sets on these issues. I suppose you are asking to what extent causal set theory realizes or enriches your perspective. Let me attempt a partial answer. To go further, the best way would probably be to refer you to specific papers, or to try to answer more specific questions if you put them to me. In a general way, it seems to me that the starting conceptions of causet theory derive in part from thoughts very similar to yours, albeit crucial inputs -- such as spatio-temporal discreteness -- come primarly from other sources. According to the conception of reality suggested by causet theory as it stands at present, "existence" would "be" an ongoing growth process consituted by "successive" births of causal set elements. Each new element is born with a definite set of other elements as ancestors (they are its "causes" or "temporal antecedents"). The particular pattern of ancestral relations that is built up among the elements represents the "causal order" that defines the causal set mathematically. Macroscopically, this pattern (or rather a portion of it) manifests itself as a spacetime manifold with a definite topology and metrical structure. [Altho I cannot make the connection concretely, I like to think that our subjective "experience of the now" is indirectly reflecting this profusion of ongoing microscopic events that (or whose "record") constitute(s) reality.] Crucially for the recovery of spacetime, the birth process in question is "asynchronous" in the sense that a given pair of births can be without direct causal/temporal connection to each other. In such a case neither element is the ancestor of the other, although they will in general have both ancestors and descendants in common. One says that the elements are "spacelike" to each other. Is this idea of an asynchronous and discrete birth-process compatible with your conception of time? Concerning how causal sets bear on the question of the "block universe", I can refer you to an article of mine entitled "Relativity-theory does not imply that the future already exists: a counterexample". See http://www.perimeterinstitute.ca/personal/rsorkin/some.papers/126.asynchronous.becoming.pdf On the same web page, the paper "Indications of causal set cosmology". http://www.perimeterinstitute.ca/personal/rsorkin/some.papers/101.peyresq.pdf offers an example where laws acquire a certain historical character. What happens in this model is that the transition probabilities governing the growth process vary from one cycle of cosmic expansion to the next in such a way that the new laws depend on certain adventitious features of the previous cycles. (This dependence follows a "meta law" that unfortunately remains "outside the process", although it takes an infinite time to reveal itself fully, and is perhaps also historical in that sense.) Could this be a rudimentary example of what you have called "transformation of transformation"? > 2. I do not see how one can dispute that the conceptual > relations among moves in logical and mathematical reasoning are > atemporal, although their thinking out is always a mental event > in time. A conclusion is simultaneous with its premise. The same > holds for the relations among mathematical propositions, even > when (as famously with the calculus) they are used to represent > events in time. I recall reading that Brouwer traced mathematical thought to a source that is not at all static: "the one splitting into two", but perhaps his was a minority opinion. In any case, I agree with you that mathematical reasoning, in its present stage of development, tends strongly toward atemporality. This tendency could change, however, and in fact one can easily notice its opposite in contemporary mathematics. One example is a renewed emphasis on algorithms in connection with the rise of computers. Indeed, in many programming languages, the equality sign `=' carries an explicitly active meaning: it stands for a process, not a logical relation. Some people also work on "temporal logic", including forms of logic in which axioms are "used up" after a certain number of uses. In set theory, axioms invoking "games" have become popular. Category theory, which some people promote as a new foundation for mathematics, also comes with "temporal undertones". In view of such developments, how can we be sure that that mathematics will not evolve to become more "temporal" in nature? Obviously, human reasoning needs a large degree of stability and reproducibility to be effective, but that's not special to mathematics. It applies equally to the kind of discussion we're engaged in right now, and I'm not sure there would be any way to avoid it. > 3. That scientists see causal links as real connections in > nature seems to me to describe only half of a > confusion. Scientists may tend to view causal connections as > real, rather than as mental constructs in the fashion of Hume > and Kant. However, they are also accustomed to regard causal > connections as instances of the regular, law-like operations of > nature. Indeed they are, but I'm having trouble convincing myself that their attitude is confused, since as I tried to describe in my last email, your distinction between laws and causal connections is not very clear to me. Possibly one should distinguish here between "temporal connections" and "causal connections", the latter being more directly bound to "lawfulness". Could one then rephrase your distinction as being between laws and "temporal connections"? Here I should add that the status of dynamical laws is something that has always confused me. > The statement of such laws in the form of a series of equations > has been the commanding project of physics. The history of > physics is, in large part, the career of those equations. These days, a theoretical talk by a physicist is more likely to begin with a Lagrangian than a set of equations. Is this significant? > However, the history of the universe being what it is, causal > connections can exist without recurring over a differentiated > structure of reality. They can therefore fail to assume > repetitious, law-like form. Such a way of thinking that cannot > be reconciled with the conventional -- and confused -- view of > causation that prevails within and outside science. > > The operational ideology of science seems to be that both the > causal connections and the laws of nature, which they supposedly > enact, are real. I argue that the causal connections are more > real than the laws. Their reality, as primitive features of > nature, is related to the reality of time. To the extent that I grasp the distinction, I tend to agree with you. (But I'm not at all sure that I do grasp it!) > 4. I disagree that the argument of which these ideas form part > is unformed or premature. There are, however, two problems. The more it's premature (if it is), the more important to keep it alive until it is able to blossom (cf. the idea of atomism). > a. As always, with the introduction of a new set of ideas, there > is no available and reliable language in which adequately to > express them. It seems unwise either to rely unreservedly on the > existing language or to invent a wholly new vocabulary. One > needs both to use and to revise the available language, in > context and cumulatively, giving new meanings to old words. Yes, that seems appropriate. > b. It is vital to develop the agenda of empirical and > experimental research to which the ideas lead. Otherwise, they > will remain impotent natural philosophy. This is exactly where I would welcome a more concrete suggestion. What sorts of empirical activity would these ideas lead to? Are there, for example, experiments which we could undertake, but which our prejudices have blinded us to? Even at the level of pure theory, where are these ideas trying to lead us? Above I sketched some instances of growth-dynamics for causal sets. Would they be examples, or do you have in mind something still more radical? If the latter, are there concrete examples of what it might be, perhaps a little piece of a new kind of mathematics or physics, even if only very lightly sketched out? Regards, Rafael