Edwin Hubble: A Life in Five Chapters

Chapter 1 · The Invented Life

A boxer, a Rhodes Scholar, and an accent

1889 – 1919 · Missouri · Chicago · Oxford · Indiana

Edwin Powell Hubble was born at Marshfield, Missouri, in November 1889, the son of an insurance executive.

He was a genuinely outstanding athlete: he set a state high-jump record, played basketball on a championship team at Chicago, and boxed. He studied mathematics and astronomy at the University of Chicago, and won a *Rhodes Scholarship* to Oxford in 1910.

His father insisted he read *law*. He did — jurisprudence at Queen's College — and hated it, and came back with a degree he never seriously used.

He also came back with an *English accent*, a cape, a pipe, and a manner of speech he kept for the rest of his life. Americans found it affected; it was.

He taught high school Spanish, physics and mathematics in Indiana for a year, coached basketball, and then went back to Chicago for a doctorate in astronomy, working at the Yerkes Observatory. He finished in 1917, enlisted immediately, and served in France — though the war ended before his unit saw significant action.

And then there is the *invention*.

Hubble told stories about himself that were not true. He said French promoters had offered him large sums to fight the world heavyweight champion. He claimed to have practised law in Kentucky, which he did not. He described war wounds and combat experience the record does not support. He said he had rescued people from drowning.

His biographer Gale Christianson documented the discrepancies extensively.

Why he did it is not clear. He was a man of real achievement who apparently could not leave it at that.

It is worth knowing partly because it is true, and partly because it makes the actual work more impressive rather than less: he did not need any of it.

“Equipped with his five senses, man explores the universe around him and calls the adventure Science.”

— Edwin Hubble, The Nature of Science (1954)

Why this matters

Hubble invented and embellished large parts of his own biography, which is documented, and the real achievements did not require any of it.

You have the boxer, the accent and the stories. What would you ask him?

Ask Hubble

• “Why did you tell stories that were not true?”
• “Why read law when you wanted astronomy?”
• “What did Oxford give you besides a degree?”
• “Did the athletics matter to how you worked?”
• “Does the invention change how we should read the science?”

Chapter 2 · How Big Is the Universe?

The Great Debate, and a genuinely open question

1920 – 1923 · Washington · Mount Wilson

In 1920 there was a genuine and unresolved question about the *size of the universe*, and it was formally debated at the Smithsonian in Washington.

Telescopes showed many *spiral nebulae* — faint fuzzy spiral-shaped objects.

*Harlow Shapley* argued they are relatively *nearby* clouds of gas *within* our own Milky Way, which is the whole universe. He had good evidence: Adriaan van Maanen claimed to have measured the *rotation* of spiral nebulae over a few years, and if a spiral is rotating fast enough to see in a decade it must be small and close, because otherwise its outer parts would exceed the speed of light.

*Heber Curtis* argued they are *separate galaxies*, enormously distant, each comparable to the Milky Way — the *island universes* hypothesis. His evidence included novae observed in spirals, which appear much fainter than novae in our own galaxy, suggesting great distance.

The debate settled nothing, because the crucial number — the *distance* — was unknown.

In 1919 Hubble joined the *Mount Wilson Observatory* in California, which had just completed the *100-inch Hooker telescope*, then by a wide margin the largest in the world.

On the night of 5 October 1923 he photographed the *Andromeda nebula*, looking for novae, and found what he took to be one. He marked the plate *N* for nova.

Comparing it with earlier plates, he found the star had *brightened and dimmed before*, in a regular cycle. It was not a nova. It was a *Cepheid variable*.

He crossed out the N and wrote *VAR!* on the plate, with an exclamation mark. The plate survives.

Because a Cepheid announces its true brightness through its period — Leavitt's relation, calibrated by Hertzsprung — he could calculate the distance.

It came out around 900,000 light years. The Milky Way was thought to be perhaps 100,000 across.

Andromeda is not in our galaxy. It is *another galaxy*, far outside it.

The universe was, in one plate, enormously larger than anyone had supposed. Van Maanen's rotation measurements turned out to be spurious.

Why this matters

One variable star on one plate settled a debate about whether the Milky Way is the whole universe, and the answer was that it is one galaxy among enormously many.

You have the crossed-out N and the exclamation mark. What is your question?

Ask Hubble

• “What was the Great Debate actually about?”
• “Why did van Maanen's rotations support Shapley?”
• “What did you write on the plate?”
• “How does one Cepheid give you a distance?”
• “How much bigger did the universe get that night?”

Chapter 3 · The Mule Driver Who Took the Spectra

Milton Humason, and the hardest observing there was

1917 – 1929 · Mount Wilson

*Milton Humason* arrived at Mount Wilson as a *mule driver*, hauling equipment and materials up the mountain during the observatory's construction. He then worked as a *janitor*.

He had left school at fourteen and had *no formal education* beyond that.

He was noticed to be careful and capable, was given work as a night assistant, and turned out to be one of the finest observers of his generation. Shapley promoted him.

What Humason did is technically extraordinary. To measure a galaxy's *red shift* you need its *spectrum*, and a distant galaxy is very faint. Getting a usable spectrum meant exposing a photographic plate for many *hours* — sometimes across *several successive nights*, with the plate stored and the telescope re-set each night.

Throughout, the telescope must track the target with high precision, and in that era tracking was *guided by hand*: the observer sits at the eyepiece in an unheated dome, in the dark, at altitude, through a winter night, making continual small corrections to keep a faint smudge exactly on the slit of the spectrograph.

For hours. For nights.

Humason took *most of the spectra* on which the redshift-distance relation rests. Hubble did the distances, largely from Cepheids and later from other indicators.

In 1929 Hubble published the result. Plotting recession velocity against distance for about twenty-four galaxies, the points fall along a *straight line*: the faster a galaxy is receding, the further away it is, in direct proportion.

That is *Hubble's law*, and the constant of proportionality is the *Hubble constant*.

The interpretation is worth stating carefully. It is *not* that we are at the centre of an explosion. If space itself is *expanding*, then every observer, anywhere, sees exactly this: everything receding, with speed proportional to distance. It is what expansion looks like from *any* point inside it.

*Vesto Slipher* at Lowell Observatory had measured the red shifts of many spiral nebulae between 1912 and 1917 — most of the ones Hubble used in the 1929 paper. Slipher is chronically under-credited, and Hubble was not always generous about it.

Why this matters

Most of the spectra behind the redshift-distance relation were taken by a former mule driver who left school at fourteen, guiding a telescope by hand for whole nights.

You have Humason at the eyepiece and the straight line. What would you ask?

Ask Hubble

• “What did taking one of those spectra involve?”
• “How does a mule driver become a great observer?”
• “Why does the straight line not put us at the centre?”
• “What did Slipher measure before you?”
• “How many galaxies were on the 1929 plot?”

Chapter 4 · A Universe Younger Than the Earth

The distance scale was badly wrong

1929 – 1950s · Mount Wilson · Palomar

There was a serious problem with the 1929 result, and it is instructive.

If galaxies are receding with speed proportional to distance, you can run the expansion *backwards*. Divide distance by speed and you get the time since everything was together — an estimate of the *age of the universe*.

Hubble's constant, on his measurements, gave about *two billion* years.

And radioactive dating of rocks already gave the *Earth* an age of *three billion* years or more.

The universe appeared to be *younger than the Earth in it*, which is not possible.

This was a real crisis and it was used as an argument *against* the expanding universe for two decades. The steady-state theory, proposed by Hoyle, Bondi and Gold in 1948, was partly motivated by it.

The error was in the *distance scale*, and it took a long time to find.

There are *two* kinds of Cepheid variable, with *different* period–luminosity relations. *Walter Baade*, working at Mount Wilson during the Second World War — when the Los Angeles blackout gave the darkest skies the observatory had ever had — realised in 1944 that the calibration mixed them up.

Correcting it *doubled* all the distances.

Further corrections followed. Some objects Hubble had taken for individual bright stars in distant galaxies turned out to be *clusters* of stars, which is a large error.

Hubble's original constant was too large by a factor of around *seven*, and the universe is correspondingly older.

The modern value gives about *13.8 billion* years, and the Earth about 4.5 billion, which is consistent.

The *relation* Hubble found was right. The *number* was badly wrong, and getting it right took most of a century — the value was still disputed by a factor of two into the 1990s, which is what the *Hubble Space Telescope's* key project was built to settle.

Why this matters

Hubble's distance scale made the universe younger than the Earth, and correcting it took most of a century — the relation was right and the number was wrong by a factor of seven.

You have the impossible age and the mixed-up Cepheids. What is your question?

Ask Hubble

• “How do you get an age of the universe from your plot?”
• “What did the age problem do to the theory?”
• “What did Baade discover about the Cepheids?”
• “How wrong was your constant?”
• “Can a relation be right when its number is wrong?”

Chapter 5 · Non-Committal to the End

The observation, and a refusal to say what it means

1929 – 1953 · Mount Wilson · Palomar · San Marino

The most curious thing about Hubble is that he never publicly endorsed the *expanding universe*.

His published position was consistently cautious: he had measured a *red shift*, and a red shift is an observation. Interpreting it as a *velocity* — and therefore as evidence of expansion — is a *theoretical* step, and he preferred to speak of an *apparent* velocity-distance relation and leave the interpretation open.

He entertained, in print, the possibility that red shifts arise from some *unknown principle of nature* rather than from motion, and he did so as late as the 1930s and after.

Whether this is admirable scientific caution or something else has been argued ever since. Some read it as a proper refusal to go beyond the data. Others note that Hubble was competitive about *priority* and may have preferred to own an *observation*, which was securely his, rather than a *theory*, which belonged to Lemaître and Friedmann.

*Georges Lemaître*, a Belgian priest and physicist, had derived the expanding solution from general relativity and predicted the velocity–distance relation in 1927 — two years before Hubble's paper — in a French-language Belgian journal that almost nobody read. He had even estimated the constant. The English translation of his paper in 1931 omitted the relevant passage, and it emerged much later that Lemaître made the omission himself, apparently thinking it superseded by Hubble's better data.

In 2018 the International Astronomical Union recommended the relation be called the *Hubble–Lemaître law*.

Hubble campaigned for years to have astronomy made eligible for the *Nobel Prize*, which was not then recognised as physics. He hired a publicist to press the case. The rules were changed to include astrophysics — in 1953, the year he died, and too late.

He was working with the new *200-inch* telescope at Palomar, the largest in the world, in his final years.

He died at San Marino, California, in September 1953, aged sixty-three, of a stroke. His wife Grace never disclosed where he was buried and there was no funeral, at his wish.

The *Hubble Space Telescope* was launched in 1990.

Why this matters

Hubble measured a red shift and refused for the rest of his life to say publicly that it means the universe is expanding.

You have the caution, Lemaître and the prize that came too late. What would you ask?

Ask Hubble

• “Why would you not say the universe is expanding?”
• “Is a red shift the same as a velocity?”
• “What did Lemaître do two years before you?”
• “Why campaign so hard for a Nobel Prize?”
• “Was your caution scientific or strategic?”

What Hubble changed

The proportionality between recession speed and distance is the observational foundation of modern cosmology and of the Big Bang model, and the constant of proportionality carries his name — now often the Hubble–Lemaître law. His 1923 plate of Andromeda expanded the known universe from one galaxy to countless ones, and his classification of galaxies by shape is still used.

A debate that continues

Georges Lemaître derived and predicted the velocity–distance relation two years before Hubble published it; Vesto Slipher measured most of the red shifts used; Hubble's distance scale was wrong by roughly a factor of seven; and he embellished substantial parts of his own biography.

Keep exploring — ask Hubble

• “How far should an observer refuse to interpret?”
• “Who else should be on the law?”
• “What would you have done with the 200-inch?”

Related lives

• Ejnar Hertzsprung — Who Found Giants Among The Dwarfs
• Cecilia Payne — Who Read What The Stars Are Made Of
• Christian Doppler — Whose Principle Rode a Railway Carriage
• Albert Einstein — Nobel Laureate · Author of Relativity

Related themes

Galaxies and the expanding universe · Red shift · Measuring cosmic distances