The Beginning of Infinity: My First Reading


  • Justified Belief vs Fallibilism
  • The scientific method is not just about being testable
  • Edison’s 99% perspiration was not mindless
  • Creationism will become an unremarkable attribute of a god
  • Genes and Memes – but not what you’re thinking about
  • The drawbacks of proportional representation
  • That Easter Islanders survived for so long is a tragedy

Why did I read this book?

Naval Ravikant – a technologist and preacher of wealth (preachings here!) – recommends this book as one of the most important to read. Naval’s approach towards reading is that he would prefer to read one hundred books many times each than read thousands of books just one time each. I think this logic makes sense. The best books/films bring about new learning/entertainment each time you read them.

What was it like for me reading the book?

Many of the chapters – such as one covering the nature of infinity, and another covering quantum entanglement – were impossible for me to understand.

David Deutsch’s primary premise for the book is that everything in the world, universe and beyond has the potential to be understood. It is a claim that can not be proven but, ultimately, no explanation can be entirely proven – it can only be checked for consistency against other explanations.

Confused by my last sentence? If so, that is exactly how I felt while reading 75% of this book. Still, there was at least 25% of the book that I understood, and Deutsch has quite a few takes (many inspired by Karl Popper, the philosopher from the late 1900s).

Here is a breakdown of what I think I have understood so far. I don’t have the intellectual background to be able to refute many of his arguments, so I present them largely as he does, providing only what minor challenges to them I can.

Justified belief vs fallibilism

When it comes to knowledge, we often refer to something being true when it is deemed true by some higher authority. This may have been a God (or high priest) proclaiming a truth of how the world was created. In more recent times it may be the World Health Organisation on how Covid-19 is spread, or the International Monetary Fund on how third world economies should run their economies. [Both are my examples, not Deutsch’s.] Historically, we had “trust the Gods”, now we sometimes have “trust the scientists” or “trust the experts”. In cases like this where truths are defined by appealing to a higher power, the approach is called justificationism.

The alternative to justificationism is fallibilism, which is the idea that there is no way to fully prove any explanation, and all explanations are error prone.

Importantly, fallibilism does not mean that we cannot explain anything, it means that we have to always expect we could be wrong. Fallibilism also does not mean that all explanations are equal. There is also a hierarchy in the quality of explanations, and that hierarchy is in how universal (or difficult to vary) an explanation is. For example, Newton’s classical theories of gravity were good, but Einstein’s are more general. I’ll touch more on this hierarchy of explanations in the next section.

The scientific method is not just about being testable

Deutsch points out that we often think about the scientific method as putting emphasis on testing our predictions. However, testability is not sufficient.

For Deutsch (and I think Karl Popper), the scientific method is about more than just being able to make predictions. For example, you might well be able to predict that a bird will reappear at the end of a magician’s trick. You may observe this repeatedly and this prediction may be borne out empirically, and others are able to test this prediction. However, this is not the scientific method, because you still may not know how the trick works. The belief that science can only provide predictions, but not explanations, is called instrumentalism and it is inherent any time we draw a graph of two variables without providing an underlying explanation. The scientific method is about explanations for causation, not just demonstrations of correlation, which is instrumentalism.

As a brief aside, the statement “this is my truth” can be interpreted as a form of instrumentalism. It takes an approach that there are no underlying explanations that can objectively be considered, there are just predictions and occurrences, and all explanations are entirely subjective and depend on your personal point of view. The scientific method, by contrast, at least according to Deutsch, requires explanation to be possible.

Zooming back out, for Deutsch, the scientific method involves explanations that are testable and lastly these explanations have a hierarchy, so not all explanations are equal. That hierarchy is determined by how difficult the explanation is to vary.

Let me start with an explanation that is easy to vary. Consider an explanation whereby global warming is explained by a God Athena, who becomes angry when humans emit carbon dioxide into the atmosphere. When we emit CO2, she causes heat to be retained close to the earth’s surface.

You might argue that the Athena theory is not testable, not least because there is no precise description of how she retains heat close to the earth’s surface. But, it is testable. You could monitor CO2 emissions and heat flows and determine that’s Athen’s prediction of a warming surface is correct. Conversely, there are aspects of what we consider good explanations that are not testable. For example, we can test the absorbing and emitting properties of carbon dioxide, but, we cannot conduct an experiment that independently varies the carbon dioxide content of the atmosphere while holding all else constant. So, what makes our greenhouse effect explanation better than the Athena explanation? The answer is that in the greenhouse theory, the details play a precise role in the explanation and are hard to vary. This is unlike Athena’s anger, which is imprecise and easy to vary as an explanation – making it inferior.

Edison’s 99% perspiration was not mindless

“Invention is 99% perspiration and 1% inspiration” is a quote commonly attributable to Thomas Edison of lightbulb and GE fame.

Deutsch argues that the 99% perspiration is in fact creative work. It typically involves human thought, not mindless work that a computer could do.

If a portion of that 99% perspiration can be done by computers it is only because something was learned by humans when they did that perspiration and programmed the computer! By definition, if something has been automated, the fact that humans can automate it means that there was learning/creativity/innovation in the work – so it wasn’t really perspiration.

In short, if a portion of perspiration can be automated, then there was learning in that perspiration!

Creationism will become an unremarkable attribute of a god

Virtual reality will become so good that we will be able to recreate a high fidelity version of life on earth. At that point, it will be unremarkable for a god to have created the earth.

Genes and Memes – no, not the kind of meme you’re thinking of

Deutsch makes the case that there are two pieces to human progress. First – genetic evolution – a slow phenomenon, and second – ideas (what he calls memes) – the more rapid phenomenon that has driven human progress over the last few thousand and, particularly, few hundred years.

Human genetics provided a platform allowing humans to develop ideas and pass those ideas down through generations as knowledge. Over the last thousand of years our genes haven’t changed so much as our memes.

Why then have humans seen such strong technological progress in the last hundred years, as compared to the last thousand. Deutsch’s argument is that we have long had the capacity to build an infinite databank of ideas, but we did not always have the culture of criticism that permitted (much less encouraged) such growth of ideas.

While critical cultures have existed for periods before (for example, during the Medici’s in Italy in the 15th century), they did not last or develop technologically as the West has done since the Englightenment.

What could some alternate theories be to Deutsch’s (and I am no expert here)?

Genetic improvement: One theory might be that humans adapted through evolution to be superior from the 17th century onwards. This seems unlikely given genetic evolution has a much longer time scale than centuries.

A triggering technological event: Another hypothesis is that we discovered the printing press, and that allowed the proliferation of ideas. A counterargument is that – while books and information were pivotal to growth – much technical progress did not occur for some centuries after Gutenberg’s press in the 15th century. There are also many other discoveries (steam engine, algebra) that we could point to as important, but it is difficult place one single discovery above all others. If we can’t pick out one single discovery, we revert to the question of why there seems to have been a collection of discoveries focused on the most recent centuries.

Geography: In the book “Guns, Germs and Steel”, Jared Diamond makes the argument that progress depended on geographical factors, such as animals being available for domestication in Europe but not so much in Africa or South America. As I see it, this seems a little bit like the “triggering technological event”, except the theory is of triggering geographical factors (presence of animals, good weather, good terrain etc.). However, these geographical factors have existed for quite some time longer than the technological progress of the last centuries, so this theory does not fully explain the question of why now? Why the progress this last century and not in the first or the fifth or the tenth century?

Back again to Deutsch’s argument that ideas drove progress and that the growth of ideas emerged from societies allowing criticism. Back in 0 BC, or even before, humans had the genetics to create knowledge as are doing now, but they didn’t. Why not? Because their societies didn’t encourage critical thinking – says Deutsch. Why then didn’t those societies allow critical thinking but later ones did? It’s not clear, and – to be fair – Deutsch is clear that this is a difficult and unanswered question.

The low hanging fruit in science and engineering is not gone

Going through a PhD in mechanical engineering and making use of equations that Sadi Carnot invented in the 18th century, I could only think that all of the low hanging fruit in science and engineering was gone.

Richard Feynman seems to have subscribed to a similar view, and felt there could only be a finite amount of theories that would soon all be discovered.

Deutsch’s philosophy is the exact opposite. Deutsch believes that there are infinite theories that remain to be discovered and so we not even close to the end. In fact, we will always be close to the beginning.

What is Deutsch’s basis for this? First of all, he points to all of those who have proclaimed the end of science before – only to be proven wrong. A good example is scientist Albert Michelson (of Michelson-Morley experiment fame) who – in 1894 – proclaimed that all physics that remained to be discovered was about the sixth decimal place. In other words, the low hanging fruit was gone and all that remained were minor tweaks to explanations and discoveries.

Second – and a rather technical point – we still do not have a reconciliation of quantum theory and general relativity. This is a clear and known gap in our knowledge of one of our most universal theories to date.

Third, it has been common in our past for theories that were widely accepted to be replaced by completely new frameworks. For example, Newton’s theory of gravity has been replaced by Einstein’s relativity. While much of Newton’s theory is a good approximation for behaviours on earth, Einstein’s theory is more universal and very different mathematically.

So, I left that chapter with more optimism than when I finished my PhD.

The Drawbacks of Proportional Representation

Today, many countries are migrating towards proportional representation – as we have in Ireland. The touted virtue of proportional representation is that it does a better job of representing minorities, which is true. Deutsch makes a number of counterpoints:

Deutsch – taking from Karl Popper – believes government should be designed such that rulers can be removed without violence if they are doing a bad job. Proportional representation means that governments are often composed of a coalition, making it difficult to remove the entire government from power (inevitably, some of the parties in power often stay on). Furthermore, proportional representation gives disproportionate power to the kingmaker parties – those that are often third or fourth in the polls and determine which larger party is in power. This means that the top polling parties are less incentivised to adapt their policies to the public, as their goals shift towards accommodating minority partners. Seeing the Irish system (and I believe the Israeli and Dutch systems are similar), I can appreciate Deutsch’s points.

What does this aside on political systems have to do with the growth of knowledge? For Deutsch, political systems should not solely be about who should be in charge, they should allow for the testing of different options that build upon knowledge, and allow for the bad options to be ditched. The question for Deutsch is not: “Is a democracy or a dictator or a monarchy better?”. The question is: “Which system allows for bad systems or policies to peacefully be cycled out?”

I won’t get into the discussion here, but Deutsch additionally makes the point that any system of representation will have logical inconsistencies, meaning that achieving “fair representation” cannot be the core goal of a political system. Improvement of knowledge can be the goal.

It was a tragedy that Easter Islanders survived for so long

I’ll finish on this example from one of Deutsch’s final book chapters.

Deutsch tells the story of two documentaries about Easter Island – an island where the civilisation went extinct in the midst of building large stone monuments.

The first is by Jacob Bronowski (commissioned by David Attenborough) and called The Ascent of Man, and it portrays a civilisation that failed in making sufficient technological progress to survive.

The second is by David Attenborough and called The State of the Planet. It draws a parallel between how the Easter Island civilisation died as it did not look after their environment, much as we today face extinction if we do not look after planet earth.

For Attenborough’s recent documentary, the extinction of the Easter Island civilisation was a tragedy. For Bronowski, the fact that the Easter Island civilisation survived for as long as it did was a traged. Whereas Attenborough sees a failure of responsibility to the planet, Bronowski sees a failure of responsibility to themselves as their population suffered needlessly building pointless stone monolith’s rather than thinking critically about how to advance their civilisation. You might think both are arguing for the same thing, but Deutsch thinks not. Attenborough is saying to go backwards to a life that is more sustainable. Bronowski is saying to go forwards and progress to the next level of technology.

In doing so, Deutsch raises the question: “What is the basis for sustainability?” Should sustainability be based upon living standards of the 1800s or the 1900s or the 2000s? Or is sustainability the wrong approach? For Deutsch, the answer is not sustainability, it is infinite progress.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s