A Brief History of Time

Stephen Hawking, 1988

Executive Summary

The book that sold millions of copies and was finished by almost none of its buyers is not, in the end, a book about the universe. It is a book about a boundary: what it means to reach the limit of a thing that contains you. Every chapter sets one up, and every one dissolves it.

Hawking’s thesis is that space and time have no far side we could ever reach. A black hole’s horizon is not a surface you bump into but a line past which signals cannot come back. The Big Bang is not a place where something was switched on but the moment the laws break down, at which point “before” stops being a word with a referent. The book’s deepest move, developed in the eighth chapter, is stated in a sentence he liked well enough to repeat: “One could say: ‘The boundary condition of the universe is that it has no boundary.’”

The method underneath the tour of physics is an epistemology. Early on he sets out the position plainly: “a theory is just a model of the universe, or a restricted part of it, and a set of rules that relate quantities in the model to observations that we make. It exists only in our minds and does not have any other reality (whatever that might mean).” Everything after that page is a demonstration of what such a model can and cannot do. The real argument of the book is that cosmology is the study of horizons, and the last horizon it comes to is the one between a model and the thing modeled — which is the horizon it finally tries to cross, and cannot.

That attempt is the famous last sentence, and it is where the book parts company with its own method. This summary will argue that the parting is the most interesting thing about it.

The Architecture, and the One Equation

Eleven chapters, 198 pages, one equation: E = mc². The count was not an accident. Simon Mitton, a Cambridge editor who read the manuscript, put the trade’s arithmetic to him directly: “Look at it this way, Steve—every equation will halve your sales.” Hawking took the advice, kept Einstein’s equation, and wrote in his acknowledgments that he hoped it would not scare off half his readers.

Three short lives follow (Einstein, Galileo, Newton) and then a glossary. The shape is a funnel. Chapters 1 to 3 build the stage, 4 to 7 replace the stage with quantum mechanics and then break it, and 8 to 11 try to close the account. Chapter 5 is the odd one: a catalogue of particles that interrupts the argument and is the least necessary chapter in the book.

The book itself has a production history worth knowing, because it explains its tone. It began after Hawking gave the Loeb Lectures at Harvard in 1982, ran to a first draft in 1984 and a final draft in the autumn of 1987, and reached American shops on April Fools’ Day 1988. The manuscript at acquisition was, in the words of the editor who bought it, “a slender but extremely dense 100 pages.” Cambridge University Press had offered £10,000; Bantam offered $250,000. Carl Sagan wrote the introduction, which appears in no edition after the first, and it opens on the sentence that describes most of the book’s readers: “We go about our daily lives understanding almost nothing of the world.”

By Hawking’s own count in the 1996 foreword, it stayed on the London Sunday Times bestseller list for 237 weeks, was translated into something like forty languages, and sold “about one copy for every 750 men, women, and children in the world.”

5 Core Arguments

1. A theory is a model, and the book means it. The sentence quoted above is not throat-clearing. It is the rule Hawking plays by for ten chapters, and it has a consequence he is unusually honest about: a model is judged by whether its predictions match observation, not by whether it resembles reality. When he later says the universe might have no boundary, or that time becomes space-like near the beginning, he is describing what the mathematics does under a condition, not reporting on furniture. Readers who skip that page read the rest of the book as a description of reality, which is exactly the mistake the book warns against on its eleventh page.

2. Every edge in the book turns out to be a horizon. The event horizon is the first: not a wall, but a zone from which no signal can return, so the inside cannot be observed from the outside. The singularity is the second: the place where the equations break, which means the place where the theory announces its own limit rather than describing a thing. Then, in chapter 8, the third: if time itself becomes space-like at the beginning, then the beginning is not an event that happened at a moment, and asking what preceded it is like asking what lies north of the North Pole. “The boundary condition of the universe is that it has no boundary” is the same move applied to existence. A boundary is not always a place. Sometimes it is the limit of the questions that have answers, and the book’s real skill is teaching a reader to feel that distinction.

3. The creator is dismissed by grammar, not by evidence. The most quoted argument in the book is not a measurement. It is this: “So long as the universe had a beginning, we could suppose it had a creator. But if the universe is really completely self-contained, having no boundary or edge, it would have neither beginning nor end: it would simply be. What place, then, for a creator?” Note what is and is not happening. He is not arguing that no god exists. He is arguing that if there is no boundary, the word “before” has no application, and a creator is a cause, and a cause requires a before. Sagan understood this in 1988 and said it more sharply in the introduction: the conclusion of the effort, at least so far, is “a universe with no edge in space, no beginning or end in time, and nothing for a Creator to do.”

4. Time’s arrow is located, not explained. Chapter 9 is the best-argued chapter in the book. The laws of physics do not distinguish forward from backward, yet the world does, and Hawking’s account runs through three arrows: “The increase of disorder or entropy with time is one example of what is called an arrow of time, something that distinguishes the past from the future, giving a direction to time. There are at least three different arrows of time.” The thermodynamic arrow is disorder increasing; the psychological arrow is memory; the cosmological arrow is expansion. His claim is that the psychological arrow reduces to the thermodynamic one, and that the whole asymmetry traces back to one fact: the universe began in an extraordinarily low-entropy state. The chapter says this clearly. What it does not do is explain why the initial state was low-entropy, and that omission is the hinge on which the book’s critics have pressed for thirty-five years.

5. The chapter the book is famous for is the one its author later said was wrong. Chapter 7, titled “Black Holes Ain’t So Black,” delivers the result Hawking is best known for: black holes radiate, they have entropy, and their area never decreases. The 1988 text then treats the loss of information behind the horizon as acceptable, and this is the claim he had to withdraw. The 2017 appendix, which is part of the book now, says so in his own words: “When I wrote A Brief History of Time, I believed that the information concerning what had fallen into the black hole was truly lost, perhaps residing in a separate universe hived off from our own. In 1997 I even bet Caltech physics professor John Preskill an encyclopedia of his choice that I was right.” And then: “It was only later, in 2004, that I realized I had been wrong.”

A Respectful Disagreement

The strongest published objection came out the same year as the book, from a reviewer who accepted nearly all of the physics. Martin Gardner, writing in The New York Review of Books, is a philosophical theist, and he attacks the ending rather than the science: “I firmly believe that it is not possible for science to discover any fact, or confirm any theory, that has the slightest bearing on why the universe bothers existing.” That is the clean version of the objection, and I think he is right about the logic while wrong about the tone: Hawking is not claiming that physics will answer the question of existence, and his own “breathes fire into the equations” passage concedes the gap. What the book claims is narrower and harder to dismiss, namely that if there is no boundary then one form of the question dissolves. Gardner’s sharper hit is about craft: “Hawking makes no attempt to explain his model except by a vague analogy.” That is true of the no-boundary proposal, which arrives as a globe and a North Pole and is never given a testable consequence in the 1988 text.

The technical objection is Huw Price’s, in Nature two years later, and it is aimed at chapter 9: “Hawking suggests that he has done it, but doesn’t explain how. Readers are entitled to feel a little dissatisfied.” His summary of the difficulty has not been improved on: “there is a tantalizing gap in Hawking’s popular account of his endeavour to explain the asymmetry of time.” The gap is real. The book asserts that the psychological arrow is the thermodynamic arrow, and it asserts that the thermodynamic arrow follows from a low-entropy start, and it never derives the low-entropy start from anything. It is a boundary condition dressed as an explanation.

The third objection is observational, and it was made the same week the book appeared. Marcia Bartusiak, reviewing it for The New York Times Book Review, accepted the physics and noted what the physics could not yet be checked against: “Current astronomical observations do not support Mr. Hawking’s vision, as yet; not enough matter, either luminous or dark, has been found to close the universe back up.” Hawking admitted as much in the text. The point is not that he was wrong to write the proposal down. It is that a proposal with no observation behind it was presented as a conclusion.

James Gleick, reviewing a biography of Hawking in 2021, attacks the destination the whole book points toward: “The theory of everything is a false idol. Why should the universe, which grows more gloriously complex the more we see, be reducible to one set of equations and formulae?” His conclusion is a direct rebuke to chapter 11: “Let us hope there will never be a final theory.”

Here is my own disagreement, and it is with the last page rather than the middle. The book spends ten chapters establishing that a theory is a model that exists only in our minds, and then closes by promising a model that would tell us the mind of God. That is not a conclusion the method can deliver, and the sentence works only because it is doing theology with the borrowed authority of physics. Carlo Rovelli, a friendly fellow physicist, put the same point gently: “although he liked to evoke God for rhetorical effect, he was resolutely atheist.” The God of this book is a rhetorical device, and the closing sentence is where the device is asked to carry an argument. It is also, by a wide margin, the most quoted sentence in the book, which is diagnostic: readers were not shopping for a theory of everything. They were shopping for a sentence about God, and the book supplied one it had not earned.

The same complaint applies to imaginary time, the technical device that makes the no-boundary proposal work. Joseph Silk, reviewing a later Hawking book in 1996, measured the gap between the exposition and the reader: “I expect that only a mere handful of the ten million odd readers of A Brief History of Time could cope with imaginary time.” The device is not hard because the reader is slow. It is hard because the book introduces it by analogy and never shows what it buys, so the reader is asked to accept a move without being given the reason for it.

That reading is supported by something harder than opinion. Jordan Ellenberg’s “Hawking Index,” which estimates how far into a bestseller readers actually get by looking at the passages they highlight, put A Brief History of Time at 6.6 percent, second on a list of the most-started and least-finished books. A book about horizons has a horizon of its own, and most of its buyers stopped well short of it.

And a last complaint about what the book does with its own uncertainty. The 1988 text presents the no-boundary proposal with a confidence that later editions do not support. In the 2017 appendix Hawking writes that the proposal’s account of the microwave background agrees with observation, and then adds: “there is one key prediction of the theory that has yet to be verified.” That is the honest state of the question, and it took a new appendix, thirty years later, to say it.

Bottom Line

Read chapters 1, 2, 3, 8, and 9. That is the book. Chapter 1 gives you the epistemology, 2 and 3 give you the stage, 8 gives you the boldest idea in it, and 9 gives you the best argument in it. Skip chapter 5 unless the particle zoo is what you came for; it is the most dated and least load-bearing chapter, and nothing later depends on it.

What has aged badly: the confidence of the ending, and the treatment of information loss as a settled consequence rather than a crisis. What has aged well: the model epistemology, and the habit of dissolving an edge into a horizon. No popular book since has explained more cleanly why “before the Big Bang” may be the wrong question, and the reason is not that Hawking knew more than his successors. It is that he understood what a question needs in order to have an answer, and he was willing to end a chapter on that.

Read the 2017 edition rather than the 1988 one, and read it for the appendix. It is the best chapter in the book and it was not in the book. There Hawking reports that inflation and the no-boundary proposal now point toward a multiverse, that gravitational waves have been detected, and that he was wrong about information. A famous author who returns to tell you which parts of his own famous argument did not survive is doing something rarer than popularizing physics.

“If we find the answer to that, it would be the ultimate triumph of human reason—for then we would know the mind of God.” — Stephen Hawking, A Brief History of Time (Bantam, 1988), final chapter, full text

Sources

• Stephen Hawking, A Brief History of Time, 2017 Bantam edition, full text (carries the 1988 text, the 1996 foreword, and the 2017 appendix)
• Stephen Hawking, A Brief History of Time, 1988 Bantam hardcover scan, full text
• Marcia Bartusiak, “What Place for a Creator?”, The New York Times Book Review, April 3, 1988
• Martin Gardner, “The Ultimate Turtle,” The New York Review of Books, June 16, 1988
• Huw Price, “A Point on the Arrow of Time,” Nature 340 (20 July 1989), 181–182 (author’s preprint)
• Joseph Silk, “Arrow-minded views,” Times Higher Education, March 15, 1996
• James Gleick, “Eclipsed by Fame,” The New York Review of Books, April 29, 2021
• Carlo Rovelli, “To Infinity and Beyond,” Foreign Policy, Winter 2019
• Michael White and John Gribbin, Stephen Hawking: A Life in Science, ch. 14 (National Academies Press, open access)
• Peter Guzzardi, “A brief history of A Brief History of Time,” The Guardian, March 14, 2018
• Simon Mitton on the writing of A Brief History of Time, BBC News, March 14, 2018
• Jordan Ellenberg’s Hawking Index, as reported by The Independent, July 2014

PRH | huffmanwrites.org | © Philip Huffman