The Imagined Humanism of Yuval Noah Harari

A review of Homo Deus: A Brief History of Tomorrow (2016), by Yuval Noah Harari.

In the film 12 Monkeys, Bruce Willis plays a prisoner who travels back in time to gather information about the outbreak of a virus that has since decimated the human population. In doing so, however, he (spoiler alert!) inadvertently sets off the very chain of events that leads to that horrific outcome.

In his 2016 book Homo Deus, Israeli historian Yuval Harari finds himself on a similarly tragic quest.

While Harari’s first book Sapiens (2014) presented a sweeping and thrilling account of the history of the human species, Homo Deus promises nothing short of a brief history of our future. The sheer scope of his work across both books is impressive — he rides roughshod over traditional academic boundaries to draw out many genuine insights across the fields of biology, history, anthropology and psychology.

Permeating both books is Harari’s keen interest in the grand ideologies that have shaped human lives, with a particular focus on ‘liberal humanism,’ which he considers currently to be the most dominant in the developed world. Perhaps the most interesting and contentious claim in Homo Deus is that this philosophical outlook is doomed to collapse, as its key premises are being eroded by the findings of modern science.

The irony is that it is actually the fallacious arguments Harari puts forward themselves that, if taken seriously, fatally undermine the core tenets of liberal humanism. And, just like inventing a toxin capable of obliterating the human race, subverting the philosophical worldview primarily responsible for the rise in human well-being over the past few centuries is one of those Very Bad Ideas.

Before addressing Harari’s critique of liberal humanism, we should first question the intuitive assertion repeated throughout Homo Deus that it is in fact the globally dominant philosophical doctrine. To the contrary, if free speech, equality before the law, and democratic elections are basic demands of liberal humanists, then they as a group still definitely only constitute a global minority. As of 2016, only roughly 40% of the world’s population were living under liberal democratic forms of government. And in case you think that figure might have been trending upward, it had actually dropped from 47% in 2007.[1]

So while liberal humanism has enjoyed a fantastic run over humanity’s recent past, it has by no means achieved global domination. It therefore still requires strong and consistent advocacy to survive — let alone thrive — in the face of serious and charismatic alternatives such as ethno-nationalism, communism, and theocracy.

With that in mind, let’s take a look at the substance of Harari’s critique.

He begins by laying the two basic assumptions behind liberal humanism. The first is that human beings consist of individual selves who possess free will, and the second is that all human beings are morally equal to each other.

It’s fairly obvious why these concepts underpin key humanist political principles like liberalism and democracy. After all, if we are not autonomous, rational individuals, why on earth should we be allowed to freely speak our minds, choose our careers, or enter contracts with each other? And if we are not moral equals, it must be truly insane to grant us all an equal vote at the ballot box.

Harari’s striking contention is that modern science is in the process of undermining both of these assumptions. Thankfully (or not, depending on how influential his ideas turn out to be) he’s wrong.

Free will and ‘the self’

Let’s start with Harari’s claim that science is showing that we lack free will. This has been a very popular provocation by contemporary scientists and philosophers (such as Sam Harris, Jerry Coyne, and Gregg Caruso) in this age of advanced genetics and brain imaging techniques such as fMRI.

Harari joins the fray by asserting that ‘free will’ was only a viable concept before scientists gained the understanding and technology required to open up the ‘black box’ of the brain. Now that scientists can observe that human choices are simply the result of genes, hormones and neurons, Harari argues, it is no longer tenable to posit such an ethereal extra property that ultimately controls our decision-making. This is because our decisions are reached through a chain reaction of biochemical events, and therefore at no point can some magical ‘freedom’ emerge to direct that course of events.

It doesn’t help that Harari fails to cite the views of any prominent philosopher or scientist on the complex issue of reconciling science with free will. This results in him unfortunately recapitulating very old fallacies that have blocked clear thinking on this issue for centuries.

Harari’s first line of argument is that we cannot observe the event of a ‘conscious free decision,’ when we crack open someone’s skull and peer into their brain. This is a true, yet utterly confused point to make. Imagine your co-worker suddenly smashing open your PC, and shouting, “See?! All that talk of your magical ‘Microsoft Word’ is nonsense! All there is in here are transistors and wires, obeying the laws of physics and computation!”

The mistake is one of fixating at the wrong level of description. Of course we cannot observe free will in action when looking at neurons under a microscope. But if we zoom out a little, to the level of people, desires and beliefs, we cannot help but observe free decisions being made all around us. It’s the very water we swim in as intentional, thinking, social beings.

The general error that Harari seems to be making is the classic fallacy of division: demanding that a property of the whole of an entity also be true of its parts. Brains cannot have free will, it is claimed, because neurons do not have free will. This would also mean however that water couldn’t be wet because the molecule H2O is not itself wet. It is similarly absurd to think that a brain cannot ‘think freely’ because the neurons that compose it cannot individually do so. ‘Free will’ is an emergent property of many, many component parts ‘doing their thing.’ One day we will understand scientifically how this thing is done, [2] but almost no one expects us to discover that individual neurons are themselves conscious and free.[3]

Harari invokes another popular argument against free will: ‘free’ decisions can’t really be free, because they can always be traced back to a series of non-conscious causes. But this reduces to the fallacious claim that any event that has a prior cause is not itself a cause. This in turn eliminates any causes from the universe except for some inexplicable first cause, and thus unhelpfully detonates any meaningful discussion of why certain things happen rather than others, the endeavour also known as ‘science’.

The reality is that different entities in this universe, at different scales, have markedly different causal properties. One of those entities is a healthy, conscious human brain, which, when suitably located in a rational and supportive culture, can begin creating knowledge that in turn transforms the matter and energy around it. Once it has done so, it is right and proper to refer to the free will of that conscious brain as the cause of those transformations.

Such a ‘will’ is indeed not actually physical in nature. Just like the software programs influencing the behaviour of your computer right now — it is an abstraction. For more on how abstractions can cause transformations in physical reality, see this chapter from David Deutsch’s excellent book, The Beginning of Infinity (2011).

Now, of course the mind cannot defy the laws of physics — it cannot be absolutely free, whatever that might mean. But if it can think creatively and critically about its own reasons for action, then it is no longer merely reactive to impinging sense data. It can autonomously plan for its future, using the resources around it to achieve its ever-updating goals.

The existence of free will is not just some inert semantic puzzle, but rather has profound ethical implications. If free will does not exist, we cannot hold people morally responsible for their actions. Most people intuitively know this, which is why they tend to be more likely to cheat or lie after reading arguments that undermine their belief in free will.[4] It is therefore unwise for Harari to so casually dismiss the existence of free will, especially when the arguments he employs already stand long refuted.

Harari goes further than denying the existence of free will, also presenting arguments that we do not even in fact possess ‘individual selves’. This is a problem for liberal humanism because it apparently presupposes that ‘we all have a single, true, authentic self, which is the source of all meaning and authority in the universe.’[5]

Harari refers to theories coming out of neuroscience and behavioural economics that the human mind consists of two different ‘selves’: the experiencing and the narrating self. The former represents our moment-to-moment awareness, and the latter is a kind of storyteller that tries to make sense of the past and plan for the future. A long discussion follows about how the narrating self uses simplistic heuristics to evaluate past experiences. One of these is the ‘peak-end’ rule, which leads us to over-value experiences if either the peak or end of the experience was particularly intense.

The stories weaved by the narrating self are usually full of contradictions and plot-holes, and Harari recounts vivid historical examples of how this can lead otherwise sensible human beings to rationalise terrible mistakes in judgement.

This certainly is interesting ground to cover, but it hardly amounts to the conclusion Harari is seeking — that we are therefore not each coherent individuals with an identity that persists over time. According to Harari, the stories that amount to our personal identity are fictions, with the same truth-status as God and heaven. But this implication is absurd.

If I tell an incomplete and flawed account of how I went to the shops, it can still be a sufficiently accurate account of reality to be taken seriously. After all, I still went to the shops. All of our beliefs contain some error, and the fact that the same holds of our beliefs about who we are shouldn’t surprise us. It certainly shouldn’t lead us consider our selves as equivalently illusory as Santa Claus.

Our thoughts do not appear to us as one perfectly consistent and coherent system of ideas. Every individual mind, by virtue of being alive, contains a cacophony of competing hypotheses about what is going on in the world and within itself. This does not mean however that there isn’t a single, unified person that this complex process amounts to. Indeed, deliberate, purposeful action is only possible if a being achieves a certain threshold unity of perception and a sufficiently ordered understanding of reality.

In short, Harari’s extraordinary claims that we lack free will and even coherent individual selves simply don’t stand up to scrutiny.

Human equality under threat?

Harari goes on to argue that future scientific advances, namely those that greatly enhance the mental and physical capacities of some humans and not others, will further erode the liberal humanist’s position that all humans are moral equals.

To make sense of this claim, we should first clarify what exactly Harari means by ‘equality’ in this context. Harari thinks that to be a liberal humanist, you need to believe that all humans, and all human experiences, are equally valuable. Once again, Harari doesn’t reference a single philosopher who has successfully advocated this position. That is because it is patently ridiculous. Of course some human beings offer more value to the world than others, and of course some human experiences are vastly more significant than others.

The real question, which has been grappled with by humanists for centuries, is what is meant by the concept of moral equality for all human beings. More specifically, is it possible to be treated as morally equal, while being substantively unequal?

A decent answer to this question is that all human persons enjoy equal rights. That is, individual human beings should be treated similarly, with any arbitrary differences (such as race / gender / sexuality) cast aside. Notice that this does not imply that we need to attribute the same value to Stephen Hawking’s life as we do to Osama bin Laden’s.

That being said, there is a profound sense in which almost all human beings belong to the same equivalence class of value: they all share a basic cognitive capacity to comprehend the truth-claims made by others, and adjust their beliefs accordingly. This shared capacity for ‘rationality’ constitutes the tremendous gulf between us and all other known forms of life. The relatively minor differences among rational human beings are dwarfed by the difference between a rational human being and a non-rational animal.

This is what many people are trying to say when they refer to all humans as being ‘equal’ — we are equally blessed with a rational nature (as Aristotle put it), despite our relatively trivial differences in intelligence, competence, and appearance.

Now that we’ve clarified what a sensible liberal humanist might mean by ‘equality,’ we can make sense of why Harari thinks his version of equality is under threat due to advances in biotechnology.

Of course, as human enhancement technologies continue to progress, humans may temporarily become substantially less equal in ability. The question is: why does this matter? Historically, all new innovations from painkillers, to vaccines, to mobile phones, start off as a privilege for the rich, but then eventually become cheap and widespread, as companies compete to sell them at a lower price and higher quality, or governments distribute them universally.

But no, it will be different this time, Harari warns us. While previous bio-technologies were aimed at healing the sick — that is, they sought to bring everyone to an egalitarian normative standard of ‘health’ — future bio-technologies will aim at enhancing human capacities, seemingly without limit. This will go on without end, with new baselines being constantly established.

But this fact doesn’t actually imply that people will become less equal over time, as the poorest will still be getting smarter and more capable as the price of the latest enhancement innovation decreases. And it will not be in the interest of ‘the elite’ to inhibit the progress of the poor, as they will need an increasingly smart, productive population to purchase their increasingly sophisticated products (including artistic creations), and to ensure people can earn the money to pay for these products.

This point belies Harari’s concern throughout the book that a vast section of the population will be left behind and become utterly valueless in the eyes of the elite. He forgets that it will be in everyone’s interests to apply these bio-technologies to everyone. To lift all boats, so to speak.

As we embark on the next phase of human enhancement, society will continue to consist of more and less competent human beings. What matters, and what liberal humanists have generally argued historically, is that we all treat each other with equal respect. This means taking each other as free, individual agents who have the right to pursue their own conception of the good, under equal protection of the law.

The meaning of liberal humanism

In his fascinating book Nothing is True and Everything is Possible (2015), Peter Pomerantsev recounts working as a television producer at the highest levels of the Russian state-controlled media in the early 2000s. At one point, he asks a member of a pro-Kremlin youth group how he defined himself politically. He replied with a smile: ‘I’m a liberal — it can mean anything.’

To land his many provocative arguments, Harari similarly needs to contort liberal humanism into an historically unrecognisable form. In doing so, he reconstructs what he thinks the core tenets of the philosophical doctrine are, and vigorously attacks them. This in turn erodes the stability of its actual meaning, which opens up breathing space for less honest and more sinister interpretations.

This brings us to perhaps the most important and persistent error in the book. This is the spurious charge that Harari consistently makes, that liberal humanism is ‘just another religion’. At one point he actually claims that medieval crusaders and modern liberals are equally delusional.[6]

To get away with this, Harari adopts the frustrating tactic of ignoring the common dictionary definition of ‘religion’: which usually requires belief in some kind of super-natural being. Instead, he weakens this definition to be a belief in ‘anything that confers superhuman legitimacy on human social structures’. He can then safely categorise liberal humanism as a religion, because its adherents clearly believe in a system of moral laws that were not invented by humans (and are thus ‘super-human’ in character).

Liberal humanists mostly do in fact hold that there are moral truths to be discovered, in particular, truths about better and worse modes of social organisation, which are not themselves arbitrarily made up by human beings.

One might say that liberal humanism assumes the existence of ‘super-human’ moral facts, that is, they endorse a kind of ethical super-humanism. This does not, however, entail epistemological super-humanism. The latter supposes that only a super-human being can truly understand which moral claims are true or false. This belief is what makes most religions distinctive — they posit the existence of a being with a mind fundamentally superior to our own, whose moral dictates in turn are ultimately beyond our comprehension. That’s why blind obedience to such beings is necessary: we have no choice but to submit to their superior wisdom, rather than come to an independent understanding of the inherent goodness of their commandments.

This is why the early liberal humanists defined themselves precisely in terms of their resistance to such blind obedience. The motto of the Royal Society when it was founded in 1663 was ‘take nobody’s word for it’. Similarly, the 18th century philosopher Immanuel Kant’s great challenge to humanity was ‘sapere aude’: dare to know.

By muddying this boundary between ethical and epistemological super-humanism, Harari can in turn claim that while factual statements belong to science, all ethical claims belong to religion. This harks back to the ‘non-overlapping magisteria’ conception of the late Stephen Jay Gould — that science and religion simply deal in different kinds of knowledge.

But a more sensible liberal humanistic account is that scientific and ethical claims are both attempts at factual claims about reality. The relevant distinction is then between rational modes of discourse about such claims, and anti-rational ones. Participants in rational discourses hold the ultimate objective of discovering and eliminating errors in their ideas (otherwise known as seeking the truth) whereas those in anti-rational discourses have a different ultimate objective — such as discovering and following the will of a superior being, or achieving some ideal political outcome or arrangement.

Harari systematically misconstrues the liberal humanist position on ethics, claiming they believe that people freely choose their values, or what constitutes the good. But they don’t tend to believe anything of the sort. They believe instead that people are able to improve upon their inherited values via freely engaging in rational discourse. That is to say, they can discover values that are objectively better than their previous ones, in the course of free inquiry. Doing so requires the capacity to rationally criticise their current beliefs, as well as to be able to imagine alternative ones.

Harari also misconstrues the liberal humanist account of politics, by claiming that it considers the individual voter to always be right. On the contrary, it is obvious to any humanist, for example, that many German voters in the 1930s were fatally mistaken. The reason liberal humanists fight for all individuals to have an equal vote is not because they are each morally infallible, or even because they are all of equal value. The reason is that it is a crude but effective means of ensuring that a minority cannot enslave the majority.

Obviously though, democracy is an extraordinarily inadequate means for ensuring that the majority does not crush a minority. That’s why we talk about liberalism and democracy as related but distinct political projects. It is possible to have a relatively liberal dictatorship, where individual economic freedom is substantial under non-democratic leadership (like 1980s Chile), as well as an illiberal democracy, where individual freedoms are restricted but majority rule is in place (like Venezuela today).

Finally, Harari misconstrues the liberal humanist account of aesthetics. Liberal humanists need not hold, as he says they must, that beauty is utterly subjective or arbitrary, and that there is no objective difference between tribal drums and a symphony by Mozart. It is possible to criticise art of all forms, and to imagine ways in which it could be better. When we hear a musical note that is not quite right in a new song, or we spot a plot-hole in the latest sci-fi movie thriller, we are engaging in rational discourse about art. Artists are genuinely trying to discover the beautiful, just as scientists are trying to discover the principles of nature, and just as political and social theorists are trying to discover better principles of social organisation.

By misrepresenting its core metaphysical and ethical claims, Harari missed an opportunity to engage with the basic message of humanism, and has inadvertently provided ammunition for alternative, aggressive ideologies. The historian A.P Norman has already taken Harari to task for grouping Nazism and Stalinism under the banner of ‘humanism’, simply because they also sought to replace the traditional authority of God with that of human beings. What matters is what we as humanity then do with that authority. Hitler and Stalin are defined by their utter dereliction of an essential humanist moral duty: to treat all individual human beings as autonomous ends-in-themselves, equally worthy of respect and dignity.

Of course Harari, like anyone, is free to use words like ‘humanist’ however he likes. But by semantically lumping together liberal humanism with other dangerous ideologies, he makes advocating for the former unnecessarily difficult.

The inevitable rise of ‘Dataism’

Toward the end of the book, Harari claims that as liberal humanism declines, a new ideology will emerge in its place, which he calls ‘Dataism.’

The exact content of ‘Dataism’ is quite hard to pin down. Harari describes it as the theory that ‘everything is data flow’.[7] A human being’s value is therefore their contribution to this data flow, presumably by how quickly one can reply to emails. The promise of Dataism is that computer algorithms will be able to understand and satisfy our desires far better than we ever could via introspection. The punch line is that the institutions of political and economic liberalism will become utterly redundant, as inferior means for achieving the end of optimal preference satisfaction.

Harari claims that Dataism is on the rise amongst Silicon Valley types. But again, he does not bother to actually reference anyone else who holds this view. This is because ‘Dataism’ is merely the dystopic realisation of his own denial of free will, individual selves, and human moral equality.

Since we are all fundamentally composed of biochemical algorithms, he claims, increasingly intelligent machines may be able to satisfy those sub-processes much better than we ever could via conscious deliberation and choice. On this point Harari is correct — if all our biochemical sub-systems do not add up to anything greater than the sum of their parts, then being enslaved by algorithms would be an excellent idea. It would be far more sensible than ‘thinking for ourselves’, whatever silly fiction that might be.

Harari concludes Homo Deus with a half-hearted plea to avoid the Dataist fate: ‘know thyself’. This is a wise suggestion, as is his advocacy of mindfulness meditation as a valuable means to do so. But it remains a strange request, given his vigorous attempts earlier to dismantle the whole concept of a self that could be known.

A better request would be to know thy humanism, without which we would be back in the dark ages, crying out to imaginary gods to save our wretched souls.

Notes

[1] These statistics are taken from Freedom in the World report by the NGO Freedom House: https://freedomhouse.org/report/freedom-world/freedom-world-2016/.

[2] For an excellent primer on the state of our current knowledge of how consciousness works, see From Bacteria to Bach and Back: The Evolution of Minds (2017) by Daniel Dennett.

[3] The minority philosophical position of ‘panpsychism’ holds that mental properties are fundamental constituents of the physical universe, while Giulio Tononi’s Integrated Information Theory entails that certain sub-components of the brain, as well as other computational devices, may possess degrees of consciousness.

[4] Vohs, K.D., Schooler, J.W. (2008). ‘The Value of Believing in Free Will: Encouraging a Belief in Determinism Increases Cheating.’ Psychological Science, 19(1), 49–54. DOI: 10.1111/j.1467–9280.2008.02045.x

[5] Homo Deus (2016), p.220. Harari fails to attribute this view to any actual liberal humanist thinker.

[6] Homo Deus, p.304.

[7] Homo Deus, p.367.