Interview: Lee Smolin and the Status of Modern Physics

lee smolinI write a science and philosophy blog called Adams’ Opticks [1], and about a year and a half ago I published an in-depth critique of Lee Smolin’s Time Reborn: From the Crisis in Physics to the Future of the Universe, a radical reappraisal of the role of “the present moment” in physics [2,3]. My article was certainly critical of the book, but also something of a labor of love, and I’m completely thrilled to say that Lee has now read the piece and would like to respond. What follows is a Q&A, with most of the questions derived from the earlier post [4].

Adam’s Opticks: Hi Lee, central to your thesis as outlined in Time Reborn and in its recent follow-up The Singular Universe and the Reality of Time: A Proposal in Natural Philosophy (co-authored with Roberto Mangabeira Unger) [5], is a rejection of the “block universe” interpretation of physics in which timeless laws of nature dictate the history of the universe from beginning to end.
Instead, you argue, all that exists is “the present moment” (which is one of a flow of moments). As such, the regularities we observe in nature must emerge from the present state of the universe as opposed to following a mysterious set of laws that exist “out there.” If this is true, you also foresee the possibility that regularities in nature may be open to forms of change and evolution.My first question is this: Does it make sense to claim that “the present moment is all that exists” if one has to qualify that statement by saying that there is also a “flow of moments?” Does the idea of a flow of time not return us to the block universe? Or at the very least to the idea that the present moment represents the frontier of an ever “growing” or “evolving” block as the cosmologist George Ellis might say?

Lee Smolin: Part of our view is that an aspect of moments, or events, is that they are generative of other moments. A moment is not a static thing, it is an aspect of a process (or visa versa) which generates new moments. The activity of time is a process by which present events bring forth or give rise to the next events.

I studied this idea together with Marina Cortes. We developed a mathematical model of causation from a thick present which we called energetic causal sets [6]. Our thought is that each moment or event may be a parent of future events. A present moment is one that has not yet exhausted or spent its capability to parent new events. There is a thick present of such events. Past events were those that exhausted their potential and so are no longer involved in the process of producing new events, they play no further role and therefore there is no reason to regard them as still existing. (So no to Ellis’s growing block universe.)

AO: Can you help me understand what you mean by a “thick present”? I’m confused because if the present moment is “thick” rather than instantaneous, and may contain events, it seems like you’re defining the present moment as a stretch of time, which looks like a contradiction in terms. Similarly, when you say that the activity of time is a process I’m left thinking that events, activities and processes are all already temporal notions, and so to account for time in those terms seems circular.

LS: I can appreciate your confusion but look, think about it this way: the world is complex. What ever it is, it contains many elements in a complicated network of relations. To say what exists is events in the present does not mean it is one thing. The present is not one simple thing, it is the whole world, therefore it contains a vast complexity and plurality. Of what? Of processes, which are dual to events.

AO: One of your main objections to the idea of eternal laws comes in the form of what you diagnose as the “Cosmological Fallacy” in physics. Your argument runs that the regularities we identify in small subsystems of the universe — laboratories mainly! — ought never to be scaled up to apply to the universe as a whole. You point out that in general we gain confidence in scientific hypotheses by running experiments again and again, and define our laws in terms of what stays the same over the course of many repetitions. But this is obviously impossible at a cosmological scale because the universe only happens once.

But what’s wrong with the idea of cautiously extrapolating from the laws we derive in the lab, and treating them as working hypotheses at the cosmological scale? If they fit the facts and find logical coherence with other parts of physics then great… if not, then they’re falsified and we can move on. As an avowed Popperian yourself, are you not committed to the idea that this is how science works?

In addition, wouldn’t the very idea of “laws that evolve and change” make science impossible? How could we ever confirm or falsify a hypothesis if, at the back of our minds, we always had to contend with the possibility that nature might be changing up on us? Don’t we achieve as much by postulating fixed laws and revising them on the basis of evidence as we might by speculating about evolving laws that would be impossible to confirm or falsify?

LS: To be clear: the Cosmological Fallacy is to scale up the methodology or paradigm of explanation, not the regularities.

Nevertheless, there are several problems with extrapolating the laws that govern small subsystems to the universe as a whole. They are discussed in great detail in the books, but in brief:

  1. Those laws require initial conditions. Normally we vary the initial conditions to test hypotheses as to the laws. But in cosmology we must test simultaneously hypotheses as to the laws andhypotheses as to the initial conditions. This weakens the adequacy of both tests, and hence weakens the falsifiability of the theory.
  2. There is no possible explanation for the choice of laws, nor for the initial conditions, within the standard framework (which we call the Newtonian paradigm).

Regarding your questions about falsifiability, one way to address them is to study specific hypotheses outlined in the books. Cosmological Natural Selection, for instance, is a hypothesis about how the laws may have changed which implies falsifiable predictions. Take the time to work out how that example works and you will have the answer to your question.

Another way to reconcile evolving laws with falsifiability is by paying attention to large hierarchies of time scales. The evolution of laws can be slow in present conditions, or only occur during extreme conditions which are infrequent. On much shorter time scales and far from extreme conditions, the laws can be assumed to be unchanging.

AO: I’m actually a big fan of Cosmological Natural Selection (which suggests that black holes may give birth to new regions of spacetime, fixing their laws and cosmological constants at the point of inception [7]) — and I can see how that is both falsifiable in itself, and would still allow for falsifiable science on shorter time scales.

Far more radical, however, is your alternative theory which you dub the Principle of Precedence. The suggestion here is that we replace the metaphysical extravagance of universal laws of nature with the more modest notion that “nature repeats itself.” The promise of this idea is that it makes sense of the success of current science whilst leaving open the possibility that truly novel experiments or situations — for which the universe has no precedent — will yield truly novel results.

To my mind, however, this notion begs many more questions than it answers. You claim, for instance, that the Principle of Precedence does away with all needless metaphysics and is itself checkable by experiment. But is it? You suggest setting up quantum experiments of such complexity that they’ve never been done before in the history of the universe and seeing if something truly novel pops out. But how could we ever tell the difference between a spontaneously generated occurrence and one that was always latent in nature and simply unexpected on the basis of our limited knowledge? And once again, as a falsificationist, shouldn’t you count the thwarting of expectations as evidence against individual theories, rather than positive proof of a deeper principle?

LS: My paper on the principle of precedence is a first proposal of a new idea. Of course it raises many questions. Of course there is much work to do. New ideas are always fragile at first.

As to how to tell the difference between a spontaneously generated occurrence and one that was always latent in nature — this is a question for the detailed experimental design. Roughly speaking, the statistics of the fluctuations of the outcomes would be different in the two cases. I fail to see how such an experiment would violate falsificationist principles.

In addition, we believe we know the laws as they apply to complex systems: they are the same laws that apply to elementary particles. To posit new laws which apply only to complex systems, and are not derivative from elementary laws, would be as radical a step as the one I propose.

AO: Can you tell me how the universe is supposed to distinguish between precedented and unprecedented situations? On the face of it, it seems like unprecedented things are happening all the time. You and I have never had this conversation before. Are we establishing a new law of nature right now, and if not, why not?

Another objection: can you tell me where novelty is supposed to come from? If the “present moment” is both the source of all regularity in the universe, and the blank slate upon which formative experiences are recorded — then what could introduce any change? Are you assuming that human consciousness and free will may be sources of genuine novelty?

LS: How nature generates unprecedented events and how precedent may build up are important questions that need to be addressed to develop the idea of precedence in nature. What I published so far is just the beginning of a new idea.

It’s intriguing to speculate about the implications for intentional and free actions on the part of living things. But in my view this is very premature. I am not assuming that consciousness is a source of novelty; I am only making a hypothesis about quantum physics. There is a very long way to go before the implications could be developed for living things.

AO: Nevertheless, it seems readily apparent from your collaborations with the social theorist Roberto Mangabeira Unger, and also the computer scientist Jaron Lanier, that you see many connections between your conception of physics and the prospects of human freedom and human flourishing. It concerns me, however, that in pursuit of a singular — very beautiful — solution to so many problems in science, philosophy, politics and our personal lives, a lot of awkward details may get overlooked.

In philosophy, for instance, you claim to show that the reality of the present moment — conceived in terms of unresolved quantum possibilities — may at last solve the problem of free will. But what of the history of compatibilism in philosophy — from David Hume to Daniel Dennett — that purports to show that our freedom as biological and psychological agents is not only compatible with the regularity of nature, but may in fact depend upon it?

LS: There are certainly common themes and influences in my work and those of Jaron Lanier and Roberto Mangabeira Unger. And I’m happy at times to indulge in some speculation about these influences. But these are very much to be distinguished from the science. The point is that I am happy to do the scientific work I can do now and trust for future generations to develop any implications for how we see ourselves in the universe. There is much serious, hard work to be done, and it will take a long time. Especially given the present confusions of actual science with the science fiction fantasies of many worlds and AI (these two ideas are expressions of the same intellectual pathology).

I agree that we have to build a counter view carefully. I don’t claim to show that my work solves the problem of free will. I suggest there may be possibilities worthy of careful development as we learn more. As for compatibilism, I am unconvinced, but I haven’t yet done the hard work needed to develop the alternative. Dan Dennett is a generous, serious and warmhearted thinker who works hard to produce arguments which are crystal clear. But talking with him or reading him, both of which are great pleasures, I sometimes find that at the climax of one of his beautifully constructed arguments, the clarity fades and there is a step which I can’t follow. I hope someday to have the time to do the hard work to convince myself whether the fault is with his reasoning or my understanding.

AO: Since I have you here, let me try to make the compatibilist objection compelling with three more questions, inspired to a great extent by Dennett’s Freedom Evolves [8]:

  1. If we turn to physics (as opposed to biology or psychology) in search of free will, are we not likely to end up granting as much free will to rocks or tables or washing machines — or indeed computers — as we do to human beings? If we are to be able to change and adapt in response to the problems we face, surely the science of free will must be the science of a human plasticity that outstrips the plasticity of nature more generally?
  2. You claim that the openness of physics may enable us to transcend the fatalism inherent in predictions from climate science, for example: in 2080 the average temperature on earth will be six degrees warmer than it is now. But what of those other predictions stemming from climate science such as: a concerted effort to reduce carbon emissions will avert disaster? If the true nature of physics undermines the certainty of the first prediction, does it not also undermine the certainty of the second?
  3. Setting yourself against a long history of thinkers who would write off the sensation of “now” as a psychological quirk incompatible with timeless physics, you go so far as to call it “the deepest clue we have as to the nature of reality.” But I wonder what you make of the innumerable psychological and neuroscientific studies that demonstrate the problematic nature of human perception of time over short intervals? Benjamin Libet’s apparent prediction of conscious decisions from unconscious brain activity seems particularly troubling. Might you be persuaded to push in the direction urged by Dennett and resist such a conclusion by arguing that an instantaneous “you” cannot be contrasted with your slow-moving brain activity, and that the search for free will and consciousness in “the present moment” is fundamentally misguided? Can we not look, instead, to the mechanically-possible processes of decision making, learning and adaption that take place over seconds, minutes, weeks and years?

LS: I don’t see why grounding human capabilities in an understanding of what we are as natural beings implies that every capability we have is shared with rocks. We have a physical understanding of metabolism, or the immune system, but rocks and tables have neither. My guess is that when we know enough to seriously address these issues, the vocabulary of concepts and principles at our disposal will be greatly enhanced compared to what we have now. Certainly we are aspects of nature and every capability we have is an aspect of the natural world.

Regarding climate change, the first is a prediction of what could happen if we don’t take action to strongly reduce GHG emissions. My point is not that the climate models are completely accurate. My point instead is that the intrinsic uncertainties in their projections are the strongest reason to act to reduce emissions so we can avert disaster however the uncertainties develop. In national defense we prepare for war because the future is uncertain. Climate change is not an environmental issue, its a national security issue and should be treated as such.

As for the objections from neuroscience, I completely fail to see the force in this kind of argument. Those studies are fascinating but I don’t think they remotely show what is claimed. Certainly the present moment is thick and the self is not instantaneous. But giving up the instantaneous moment for the thick and active or generative present (as I sketched above) does not imply that consciousness or time or becoming are illusions.

AO: Lee Smolin — thank you!


[1] Adam’s Opticks.

[2] Time Reborn: From the Crisis in Physics to the Future of the Universe, by L. Smolin, Houghton Mifflin Harcourt, 2013.

[3] For a good introduction to the basic ideas of Time Reborn, see this video.

[4] On Time Reborn as modern myth: Why Lee Smolin may be right about physics (but probably wrong about free will, consciousness, computers and the limits of knowledge), by Joe Boswell, 5 October 2013

[5] The Singular Universe and the Reality of Time: A Proposal in Natural Philosophy, by R.M. Unger and L. Smolin, Cambridge University Press, 2014.

[6] The Universe as a Process of Unique Events, by M. Cortês and L. Smolin,, 24 July 2013.

[7] For a more critical take on the idea of cosmological natural selection see: Is Cosmological Natural Selection an example of extended Darwinism?, by M. Pigliucci, Rationally Speaking, 7 September 2012.

[8] Freedom Evolves, by D.C. Dennett, Viking Adult, 2003.


Joe Boswell is a writer and a musician trying to figure out how to make a living in a world where words and music are free. He has a degree in English literature, but having learned to bluff philosophy by listening to lots of podcasts, he enjoys picking fights with eminent scientists and philosophers on his blog, Adams Opticks ( His songs are available on Bandcamp ( He does Twitter too (

Lee Smolin is is an American theoretical physicist, a faculty member at the Perimeter Institute for Theoretical Physics, an adjunct professor of Physics at the University of Waterloo and a member of the graduate faculty of the Philosophy department at the University of Toronto.

This post originally appeared here. Republished under creative commons license.


You may also like...