Time and Chance

by David Z. Albert

This book is an attempt to get to the bottom of an acute and perennial tension between our best scientific pictures of the fundamental physical structure of the world and our everyday empirical experience of it. The trouble is about the direction of time. The situation (very briefly) is that it is a consequence of almost every one of those fundamental scientific pictures--and that it is at the same time radically at odds with our common sense--that whatever can happen can just as naturally happen backwards.

Albert provides an unprecedentedly clear, lively, and systematic new account--in the context of a Newtonian-Mechanical picture of the world--of the ultimate origins of the statistical regularities we see around us, of the temporal irreversibility of the Second Law of Thermodynamics, of the asymmetries in our epistemic access to the past and the future, and of our conviction that by acting now we can affect the future but not the past. Then, in the final section of the book, he generalizes the Newtonian picture to the quantum-mechanical case and (most interestingly) suggests a very deep potential connection between the problem of the direction of time and the quantum-mechanical measurement problem.

The book aims to be both an original contribution to the present scientific and philosophical understanding of these matters at the most advanced level, and something in the nature of an elementary textbook on the subject accessible to interested high-school students.

Genres: Philosophy, Physics, Science, Nonfiction

186 pages, Paperback

First published January 29, 2001

Reviews

[Poiq Wuy · Rating 5 out of 5]

Resumen
Es un libro con un contenido maravilloso, escrito en un estilo un tanto peculiar que no llega a molestar mucho. Se trata de un análisis conciso (demasiado conciso en ocasiones) y riguroso (mucho) del problema de la flecha del tiempo. El autor está atento a las premisas de las que parten las teorías físicas para discriminar en cómo es posible que de teorías microscópicas casi-simétricas temporalmente emerja una experiencia macroscópica asimétrica temporalmente. Su propuesta es lo que luego se llamó mentaculus, que consta de tres ingredientes:

las leyes dinámicas microscópicas.
una distribución de probabilidad sobre microestados posibles, entendida como objetiva.
la hipótesis del pasado.
Leyes microscópicas
Las leyes físicas microscópicas de las teorías de los últimos siglos no son temporalmente simétricas, pero casi (es necesario invertir algunas cantidades, como los campos magnéticos, para completar la simetría).

De hecho, si los estados (configuraciones) son también las condiciones dinámicas (espacio de fases) y las leyes dinámicas son temporalmente simétricas, entonces evolución pasada y futura han de ser simétricas. Si además las leyes son invariantes bajo traslación temporal, entonces el sistema ha de ser estático.

La única manera de escapar a esto es que las condiciones dinámicas rompan esta simetría (como en mecánica clásica newtoniana, con sus segundas derivadas) o que las leyes dinámicas la rompan (como en mecánica hamiltoniana o mecánica cuántica).

Dicho de otro modo, ¿qué cambia al cambiar ? ¿cambia el espacio de fases ()? ¿cambian las leyes dinámicas ()? Si hay invarancia bajo traslación temporal, algo tiene que cambiar para que el sistema no sea estático.

Flecha del tiempo
Hay al menos tres flechas del tiempo:

Física-macroscópica (los procesos físicos macroscópicos no son reversibles)
Epistémica (nuestro conocimiento del pasado es cualitativamente distinto al del futuro)
Agencial (nuestra capacidad de actuar sobre el futuro es distinta que sobre el pasado).
Si aceptamos que las leyes microscópicas no poseen estas flechas del tiempo, y que son deterministas, ¿es posible dar cuenta de ellas?

El determinismo es importante porque nos permite colapsar dos nociones de asimetría en una: que no sepamos predecir el futuro igual que el pasado, y que los procesos físicos que pueden ocurrir en una dirección temporal no puedan ocurrir en la opuesta.

Demonios de Maxwell
Distingue entre la entropía termodinámica (referida a los estados micro compatibles con las restricciones macro) y la entropía informacional (la llamo yo así, la entropía de Gibbs, referida a funciones de probabilidad sobre microestados sin referencia a los estados macro).

Un resultado de esto es que es posible concebir mecanismos que reducen la entropía total de un sistema aislado en un cierto intervalo: el demonio de Maxwell es posible, y la segunda ley de la termodinámica no es universalmente cierta.

Ver:: https://desfondes.substack.com/p/los-...

Hipótesis del pasado, retrodicción, trazas

Postula la hipótesis del pasado como única manera de explicar que hay métodos de inferencia exitosos para el pasado que para el futuro no están disponibles: las trazas. Podemos predecir el futuro en base al estado físico presente de un sistema. De la misma manera podemos retrodecir el pasado, ambas operaciones son simétricas. Pero con el pasado tenemos disponible una operación que con el futuro no: podemos leer registros, inferir datos de estados pasados a partir de memorias, grabaciones, huellas, etc. La única forma de explicar que esto es posible es aceptando que en el pasado la entropía fue menor, de manera universal y en todo tiempo. El estado inicial del universo, por tanto, debió ser uno de baja entropía.

Cuando inferimos algo sobre el pasado mediante retrodicción, estamos pasando de un estado presente a uno pasado. Sin embargo, cuando empleamos trazas estamos haciendo uso de tres momentos: el de preparación, el inferido, y el presente. El momento inferido siempre tiene lugar entre los otros dos, toda inferencia exige un momento previo a ella de menor entropía. A partir de su huella (presente), no puedo inferir que un electrón chocó contra la pantalla (inferido) si no doy por hecho que había anteriormente (preparación) una pantalla lista para grabar el impacto.

De esta manera, reduce la flecha epistémica a la flecha termodinámica del tiempo.

Si hubiese una hipótesis del futuro, entonces seríamos capaces de predecir en base a trazas del futuro. Eso es algo que no es posible en el presente, pero no podemos estar seguros de que en el futuro no esté disponible: quizá la entropía aumenta hasta cierto punto y luego comienza a disminuir para ser compatible con un momento futuro de baja entropía.

Libre albedrío
Albert también reduce la flecha agencial a la flecha termodinámica del tiempo. Para ello considera las situaciones contrafácticas. No es necesario negar el determinismo para afirmar mi agencia sobre el futuro, basta con considerar cuáles serían mis predicciones sobre el futuro en caso de que hiciese tal o cual cosa. Esto solo funciona porque del futuro no hay registros. Las trazas del pasado impiden considerar como cambian mis retrodicciones si modifico mi comportamiento, ya que tengo acceso por otra vía al pasado, y es una vía más exhaustiva, más exitosa epistémicamente. El futuro depende contrafácticamente del presente, el pasado no. De esta manera la flecha agencial se reduce a la epistémica, y por tanto a la termodinámica.

Rel:: 2024 - The Open Universe_ Totality, Self-reference and Time - Jenann Ismael

Mecánica cuántica
Dedica el último capítulo a mecánica cuántica. Lo que le interesa es que en modelos cuánticos como la teoría GRW · Ghirardi-Rimini-Weber, se introducen fluctuaciones estadísticas a modo de ley dinámica. Esto permitiría trazar el origen de las leyes estadísticas en la mecánica cuántica. Si la teoría es cierta, las leyes fundamentales no son deterministas, y los problemas de demonios de Maxwell posibles que describió desaparecen. La segunda ley de la termodinámica vuelve a ser una ley física. Si lo entiendo bien, el postulado estadístico del mentáculus, la distribución de probabilidades en espacio de fases, resulta prescindible: son las propias leyes dinámicas las que van generando una «difusión» probabilística de los estados en el espacio de fases, y con eso es suficiente para recuperar la mecánica estadística clásica (haciendo uso, eso sí, de la hipótesis del pasado).

[Eliot Long · Rating 4 out of 5]

A super thought-provoking and in-depth book about statistical mechanics and the arrow of time (and even a bit of quantum mechanics in the past chapter). This book might've been a bit advanced for me, and I know that some nuances/details went over my head, but if you're patient with it and are willing to go back and re-read some pages here and there, it's super rewarding. Albert gets pretty deep in the subject, so it's nice to have some prior knowledge of things like phase space and basic thermodynamics, and depending on your level of knowledge you might have to look some things up.

Difficulty at parts aside, Albert's style of discussion through propositions and refutations really helps to lead the reader through such a deep field. There aren't too many definitive answers, but that doesn't make the takeaways any less significant; you'll still understand the topic better even if there's no conclusive solution. Treat this book as a mix of philosophy and physics, since that's what it is at the core (the author is a philosophy professor).

Some difficulty does come from the author's confusing style of writing (lots of more or less unnecessary asides that interrupt the flow) but you can ignore them for the most part, though that was one of the book's biggest flaws.

[Erica]

so far, so good!

[Paul Sand · Rating 1 out of 5]

[Imported automatically from my blog. Some formatting there may not have translated here.]

I don't remember why I put this book on my get-at-library list. (I really should keep notes.) And (as it happens) my current mode of library-interaction (as it were) is to check a book's availability online (or get it through Interlibrary Loan), and my physical presence onsite (as it were) is a quick in-and-out. Browsing (as I might have done pre-Covid) is discouraged.

The above paragraph apes the style of the author of Time and Chance, David Z. Albert. Lots of random parenthetical asides that don't add much information, lots of random italics. The book reads as if it were a transcript of chatty oral presentations of a particularly animated and eccentric sort. And it can get pretty impenetrable at times, by which I mean nearly always. Here's a sample paragraph from page 62:

I’ve been talking about the postulate about statistics up to now as if it more or less amounted to a stipulation that what you ought to suppose, for purposes of predicting a system's future behavior, if you are given only the information that the system initially satisfies X, is that the system is as likely to be in any one of the microconditions compatible with X at the initial time in question as it is to be in any other one of the microconditions compatible with X at the initial time in question. That’s more or less what the postulate amounts to (I think) in the imaginations of most physicists. And that (to be sure) has a supremely innocent ring to it. It sounds very much—when you first hear it—as it is instructing you to do nothing more than attend very carefully to what you mean, to what you are saying, when you say that all you know of the system at the time in question is X. It sounds very much as if it is doing nothing more than reminding you that what you are saying when you say something like that is that X is the case at the time in question, and (moreover) that you have no more reason for believing that the system is in any particular one of` the microconditions compatible with X at the time in question than you have for believing that it is in any other particular one of the microconditions compatible with X at the time in question, that (insofar as you know, at the time in question) nothing favors any particular one of those microconditions over any particular other one of them, that (in other words) the probability of any particular one of those microconditions obtaining at the time in question, given the information you have, is equal to the probability of any particular other one of them obtaining at the time in question.

Five sentences, and that last one is a doozy. And to make matters worse, the very next paragraph begins: "This is all wrong, however." Darn!

I was very much in "I looked at every page" mode for large swaths of the book. I would flunk badly if quizzed on its details. It's a slim book, I tried to tackle a mere ten pages/day, but…

Anyway: Albert has his Ph.D. in theoretical physics but moved over to the philosophy department at Columbia. The book deals with time's arrow, or: exactly how do we distinguish the past from the future?

The problem being that many (but not all) of the physical laws of the universe are invariant under time-reversal. For example, if you had a movie of the planets revolving around the sun, then played that movie backward, the planets would still seem to be obeying, blissfully, the same Newtonian laws of motion. Similarly for gas molecules in a box: they bounce off each other, and the walls of the box, elastically, and they would appear to do the same thing in a time-reversed movie. You couldn't really tell whether you were watching the movie backwards or forwards.

Fine, but that's completely at odds with our everyday experience. We can nearly always tell when a movie's running backward: when we dominoes spontaneously rising into a complex pattern instead of falling, gases collapsing into a corner of a box instead of expanding to fill the available volume, stars sucking up light, instead of emitting it, etc. "That ain't right."

So the book immediately gets into matters of thermodynamics, entropy, and statistical mechanics. But Albert notes that the underpinnings of those fields and concepts are epistemologically shaky, and attempts to firm them up. And he may do so, but don't ask me.

In the latter parts of the book, he brings in quantum mechanics, which may help things. He discusses several interpretations, and holds up one for special attention: the Ghirardi–Rimini–Weber (GRW) theory. I think that's pretty obscure, but not obscure enough to lack a Wikipedia entry.

So: my bad. If I ever had the physics chops to follow Albert's argument, they're gone now. And in the future, I'll try to have a solid Plan B in place when getting a library book.

[zimu · Rating 5 out of 5]

The last time I read one of David's books cover to cover was in sophomore year of college– this was almost two years ago. His book Quantum Mechanics and Experience made me fall in love with the field and since then, despite talking to him almost every week, I had forgotten (at times) the way he wrote. But this book, as with all of his works really, is just written in such a way where one is reminded of his passion, warmth, and love for the subject. By the end of it, you really feel like you've just been told this super cool crazy thing by a really cool friend of yours and you can't wait to share it with everyone else– the excitement is infectious, to say the least. It's a privilege to read this sort of stuff, and an even bigger privilege to have become friends with him in the past few years.

[Parker · Rating 3 out of 5]

Perhaps the only thing I learned that I’m sure will stick with me (as it were) is Albert’s explanation of the Past Hypothesis. How, to make predictions about the future, the mathematical formalism of statistical mechanics requires the postulation that entropy was lowest at the beginning of time.

I wish it was explained in more detail the implications of the fact that entropy is one of many seemingly un-objective ways of organizing the world. This is on my mind after reading about thermal time in loop quantum gravity.

At times I found Albert to be too repetitive (so graspable that it was boring) and at other times I found Albert to fail to explain what was so novel or groundbreaking about his approach compared to the textbooks.

[Lauren Romo · Rating 2 out of 5]

God this book is terribly written. I don’t even know if it was edited or if someone was just listening to Albert ramble and copied down everything he said. You cannot go two words without some sort of useless parenthetical insertion. It actually upsets me because his ideas are so interesting but his writing is atrocious.

[Pierrot Seban · Rating 5 out of 5]

Quite an extraordinary book. I don't share many of the author's premises, and I had to consciously remind myself of his methodological postulates regularly to begin to make sense of his arguments, but the book is extraordinarily efficient, rigorous and clear. As a newcomer in the field I am anxious to see what the reception and critics were.

[Muhammad Shemyal Nisar · Rating 4 out of 5]

A very interesting but at the same time very dense book.

[lucas · Rating 4 out of 5]

albert's not easy to read, but he's an incredibly precise and original thinker with respect to both the arrow of time and his other work in quantum mechanics.

[Mark Moon · Rating 5 out of 5]

This was the clearest non-technical writing about time I've ever read.