Book Review: "Darwin's Unfinished Symphony" by Kevin Laland

Biologist Kevin Laland has a gift for clear, evocative writing and clever titles. An "unfinished symphony" is a powerful metaphor, hinting at the lost works of departed geniuses. Perhaps no post-Enlightenment figure has had as profound impact on how humanity sees itself than Charles Darwin, and the chance to extend the modern evolutionary synthesis to incorporate an account of the origins of human culture and intelligence creates possibilities that will be euphoric for some readers and terrifying for others. Laland's book contributes to an emerging corpus of recent literature aiming to close the enormous gap that has opened up between the biological and social sciences.

Laland is not a historian, and those looking for a compelling account of Darwin's research on the origins of humanity should look elsewhere. "On the Origin of the Species" is famously silent on the question of human origins, and Darwin's later works on the subject ("The Descent of Man" and "The Expression of the Emotions in Man and Animals") are less well known and more heavily weighed upon by the passage of time. For Laland, Darwin's 'unfinished symphony' was the task of explaining humanity's unique mental, cultural and moral capabilities in light of evolution, and the subtitle of his book ("How Culture Made the Human Mind") gives the game away in terms of where Laland suggests the answers will be found. This is also the story of a thirty year effort by Laland and his team to find those answers: an ode to grad students and published research most of the public have never heard of. 

The boundary between biological and social evolution is fiercely patrolled on both sides, with many biologists stressing that evolution only applies to genetically-encoded behaviours (c.f. Dawkins) and many sociologists epistemologically wedded to the unnatural uniqueness of human culture. We can perhaps refer, as Laland does, to cultural evolutionary studies pioneered by Boyd & Richerson among others, which has been a remarkably successful effort to apply evolutionary tools to the study of culture. Laland, however, goes beyond that to write a book focused on gene-culture coevolution. That is, a study of how genes have influenced culture and culture, in turn, has influenced our genes. Evolutionary psychologists and behavioural economists have made strides (often flawed) into measuring biological influences on human behaviour, in the process contributing to a resurgence of naturalist ethical ideologies. Laland's position is more nuanced: our biological intuitions and mental capabilities have developed in a feedback loop to meet the needs of humanity's complex learning behaviours. 

Innovating and Copying

As a mechanism which develops and spreads new traits through a population, culture relies on the innovation of new behaviours (beyond what is encoded in a creature's genes) and the copying of that behaviour by others (which allows it to spread faster than biological transmission alone would allow). Social learning is innate and ubiquitous among humans and - it turns out - not entirely unknown amongst animals. Crucially, Laland is able to demonstrate through both mathematical models and empirical studies that what matters to the cultural capacity to accumulate fitness-enhancing improvements over time is not just copying but copying efficiently, faithfully and strategically. Learning heuristics such as imitating successful individuals or conforming with the behaviour of the majority ensure that useful information is retained over generations, and it's this capacity for strategic or contingent learning that Laland argues lies at the root of the divergence between humanity and other animals. 

Laland combines agent-based computational models of social learning with empirical studies of social learning in animals, most amusingly fish. Throughout the early chapters of the book, he builds a compelling case for why social learning is not ubiquitous in nature, noting that vulnerability to predation and the demands of reproduction make some species (and some individuals within species) more reliant on social learning than others. In Laland's observations, it's often lower-status individuals, females and children that benefit most from openness to new experiences and social learning. A certain conservative Canadian psychologist has recently invited derision by looking for the origins of social hierarchies amongst lobsters. Laland's fish studies are different because his team has gone to the effort of first proving that social learning is a mathematical solution to a general category of problem: humans are 'like fish', in this sense, because both evolved similar strategies to solve analogous problems, not because we share some biochemistry. A more sophisticated version of Jordan Peterson might be able to do the modelling to show that hierarchies are a valid solution to the problem of cooperation - but unlike Laland, Peterson simply hasn't done his homework. 

More than just the evolution of intelligence

Unlike other pop science authors, Laland's descriptions of his statistical methods and techniques are always clear and precise - we always know how much confidence we should have in Laland's conclusions and how they're derived. Where that raised an eyebrow, however, was when Laland moved away from discussing social learning to the evolution of intelligence. Laland relies on a variant of factor analysis to argue for the existence of what he calls the "primate g", a measure of the general intelligence of primates which he argues has increased alongside the demands for increasing social learning. In this way, Laland aligns himself with the view that intelligence is an all-purpose tool that has continually improved throughout human evolution, rather than an agglomeration of modules and capabilities which arose at different points in our species' history in response to different environmental and social requirements. 

Ultimately, Laland's efforts to downplay the 'social brain' hypothesis are unconvincing and I would recommend Robin Dunbar on these topics. Is a 3cm long fish that finds a novel solution to a food maze really 'innovating', or it merely following instincts that drive hungry individuals to take greater risks? It's likely that general intelligence is a poor tool for solving most social games among humans, at least,  (cf. the ultimatum game) and IQ is a poorer predictor of social status than modular personality traits. The role of emotion and motivated action is missing from Laland's account, an omission that's curious given that Darwin himself intuited that the evolutionary function of emotions was crucial to understanding the origins of man. If culture made the human mind, as Laland argues, then perhaps we should also explore the role of emotions, biases and irrationality in cultural evolution - and not just intelligence. 

In later chapters of the book, Laland offers some unsupported hypotheses in an effort to bridge the gap between primate intelligence and human symbolic expression. In his otherwise friendly review of the book, cultural evolutionary scholar Joe Henrich worries that Laland overplays the role of teaching and language in the co-evolution of culture. Laland argues that teaching is the only explanation that can account for not only the increasing fidelity of cultural transmission in humans but also the (somewhat self-imposed) requirement that proto-languages evolved transmit information faithfully and without deception. We'd have to do the modelling to find out which account is more plausible, but given the robust evidence from behavioural economics that deception and public self-justification plays a crucial role in social games the 'social brain' hypothesis remains for me the persuasive position. 

What really disappointed me about these later chapters is Laland's seeming lack of engagement with the paleoanthropological record: he never specifies where in the human family tree he thinks these capabilities first evolved; he doesn't engage with the robust debate on the physiological limitations on the language capabilities of our near-relatives; nor with the fact that symbolic culture and exponential cultural progress has only really been a feature of our species on the timescale of tens of thousands of years or so. Teaching, language, rationality and the arts are all highly important to modern human societies, but an account of how these capabilities evolved needs to bootstrap them on to other functions that plausibly existed in earlier members of our lineage that definitively lacked them. 

Enter the debate

"Darwin's Unfinished Symphony: How Culture Made the Human Mind" is exactly the sort of book we should embrace as part of a second wave of pop science explainers of gene-culture evolution. Laland takes the extended evolutionary synthesis and cultural evolution as a given, and uses those theories to motivate and explain his own perspectives on debates within the field. Like Darwin himself, Laland may ultimately be proven right on some points and wrong on others: what's important is that readers are exposed to these issues in ways that normalise the language and perspectives of gene-culture coevolution. Laland's earlier work focused on the phenomenon of niche construction, in which a species influences its environment and thereby changes the selection pressures acting upon it. As a social species, the social environment is undoubtedly the niche that humans have created for our own evolution - and it's that control over our own environment that frees us from biological determinism like that espoused by Peterson, Dawkins and Pinker. It opens minds to the possibility that we - as a species - are not done with our social evolution just yet. 

It's OK to be hypocritical (Part 2): Vegans, morality and aesthetics

In a popular post I wrote back in February contrasting individual preference and political ethics, I argued that we should all be comfortable being a little hypocritical sometimes. Having a preference for sparkling wines, for example, doesn't automatically make one a bad socialist; having inegalitarian sexual preferences doesn't render those behaviours acceptable outside the bedroom; and benefitting from privilege doesn't make one unqualified to structurally critique it. In following-up that post, I'm going to look at another YouTube discussion by streamer Destiny, this time on the ethics of veganism. I enjoy Destiny's way of dissecting topics, which is leaps and bounds more sophisticated than that of other talking heads. In short, I agree with him that given the minimal recognition of all human beings as a party to the social contract, the consideration of other entities as ethical subjects is a mere aesthetic preference from which no universal principles can be drawn. 

For the record, I am not a vegetarian nor do I think veganism is a a superior ethical position that meat-eaters are simply being hypocritical about. I took a (mandatory) course in 'vegetarian politics' as an undergrad, which did instill in me a certain distaste for meat. But I still consume it, and don't see a problem in everyone else doing the same. Nor is my view on this informed by consequentialist concerns: I get that Western diets can have adverse health consequences and that industrial agricultural has deleterious effects on the natural environment. Switching to veganism may have positive individual and societal outcomes, and I am all for stronger regulation of the agricultural sector, but I don't believe that universal ethical "oughts" can be derived from this sort of utilitarian calculus (which in any event always admits of exceptions). Instead, this particular blog is aimed squarely at "ethical vegans", who hold that it is wrong in se to exploit animals in ways we don't exploit humans. 

It's art all the way down

As a moral skeptic, I generally take the position that ethical rules are social facts, not scientific ones. A behaviour or belief can only be considered right or wrong by its congruence with the norms, rules and institutions that constitute a community of interacting individuals. Social contract liberalism is a universalist ideology that holds that the relevant community is all human individuals, although the 'thickness' of its ethical rules may vary depending on nation-state membership. One of liberalism's key philosophical difficulties, and its central contestation with ethical veganism, is why it reifies humanity as the criterion of social membership, or in other words, how we decide who is and isn't entitled to recognition as "human". The more liberalism moves away from its Christian 'natural law' roots (i.e. ' individuals endowed by their Creator'), the more open to re-interpretation this principle becomes. 

Chapter 16 of my book, "Politics for the New Dark Age" is all about the distinction between ethics and aesthetics. Aesthetics is about creating identity through the satisfaction of preferences; political philosophy is the normative ethics of collective decision-making. So long as two individuals identify as members of a shared community, they can (in principle) debate the right and wrong of a behaviour using a shared framework. And yet, as the invocation of a shared identity suggests, there are always aesthetic choices at the core: what is my identity, what is my community, and what are its ethical rules? That doesn't mean we can't argue about those choices, but such debates boil down to art criticism: we can discuss the artistic merits of "Boss Baby", for example, without any paticular view being morally wrong. 

So here's my central argument: right and wrong are ethically meaningful social statements within a the social context defined by universal liberalism. However beyond the minimal liberal requirement that we recognise of all humans as persons, the selection of decision-rules about the ethical treatment of non-human objects is an aesthetic preference from which no universal moral facts follow. Thus, while vegans, or Hindus, may elect to see cattle as moral subjects, they cannot ethically obligate other individuals and cultures to do likewise. Personally, I'm wholly comfortable with speciesism ("humanism") as a general rule, using the legal and biological category of 'human' as the threshold for moral consideration. It's a solid baseline that prevents abuse in most edge cases, meets the requirements of liberalism and satisfies our day-to-day pragmatic needs for ethical calculation.

Can we derive a "non-speciesist" rule for ethical consideration? Probably . . . . 

Destiny defines his decision rule as "social contract reciprocity": the idea that to be part of a society, and therefore subject to ethical consideration, a category of thing has to be capable, under ideal developmental conditions, of recognising and being recognised as performing the behaviours necessary to establish social trust. Destiny may underplay it, but this is very solid philosophical ground to stand on. In evolutionary game theory, there are multiple ways of overcoming social dilemmas and generating reciprocity, including kinship, reputation and identity/categorical markers. The first two generate what philosophers call special moral obligations (for example, to family and to contract partners), whereas the third is important to Rawlsian liberals, constructivists and cultural evolutionary scholars alike. 

In this framework, we exercise indirect or altruistic reciprocity (trust without expectation of direct reward) towards those who perform the behaviours we have been socialised to expect a member of the community in good standing to perform, and who in turn recognise and respond positively to our own performance of those behaviours - even though we are unrelated to them and don't know them personally. This is called 'prosociality'. The mutual construction of society on the basis of altruistic reciprocity helps explain why sociopaths are such as widespread figures of cultural anxiety, and also why systems of punishment are so heavily weighted towards ostracism, banishment and imprisonment of non-conformists. Mutual recognition as beings worthy of ethical consideration also serves as a universal "Turing Test": we cannot ever directly observe the moral consciousness of another, only react to behaviourial signals which increase or decrease our belief that it exists. 

Some hypotheticals: Violence against animals, Neanderthals and AI

Let's use violence as a case study. If we follow Jonathan Haidt and assume that moral reasoning is modular and domain-specific, we can start with the hypothesis that aversion to violence and harm ("suffering") is perhaps the most widely shared human ethical trigger. If non-human entities are not protected from arbitrary and selfish violence, then they also aren't entitled to any other form of ethical consideration. For the record, this is where ethical vegans go wrong: as virtue ethicists, their insistence that suffering is always incorrect is hopelessly naive and drives them into a rabbit hole of deciding what entities do or do not feel suffering. Consequentialists tell us that harm is sometimes productive and therefore ethical; social contract deontologists say that harm can be ethical if it's controlled by rules - the most important contemporary rule is the prohibition on directly harming another party to the social contract. The question we must ask is: is it ethically wrong to harm an entity that belongs to a category of thing that is prima facie incapable of moral reciprocation? 

First the goods news: as social animals, humans seem psychologically pre-disposed to making favourable intuitive inferences about the agency of others: we readily anthropomorphise other entities, and ancient peoples inferred agency to the land and the weather and tried to make moral bargains with the gods and goddesses thereof. So the burden of proof is in favour of inclusion: other entities face a low bar to establish that they are moral subjects. 

Now for the bad: almost all animals fail to reach this threshold. Animal behaviour is instinctive: a prey animal is incapable of recognising that a strange human is not a threat, and a hungry predator will always see a human as a potential food source. While we can override the instincts of individual animals through direct incentives (i.e. regular food provision) this is direct, rather than indirect or categorical, reciprocity. Selfish restraint ("this human feeds me") is not general moral consideration. Now, there may be exceptions amongst higher order social animals: our closest ape relatives, elephants and domesticated dogs (whose social reasoning we have irrevocably altered through domestication) for exampel. These animals appear to be capable of recognising humans as being subjects of ethical consideration unprompted, even though under usual conditions they are primarily kin-centric. But any more generous aesthetic rule is also likely to be prima facie culturally and environmentally contingent: a costly preference for expanded ethical inclusion akin to having the resources to prefer fine wines. 

A more interesting hypothetical arises in the case of our Neanderthal and Denisovan cousins, who were not homo sapiens strictu sensu but closely related enough genetically that we could still interbreed. Robin Dunbar and others have argued that the capacity for symbolic communication is what makes human organisation at scale possible. It remains an open scientific question whether Neanderthals and Denisovans also possessed this capability, and to what degree. The evidence for Neanderthal symbolic culture is anecdotal, but personally fairly convincing. It's appears likely to me that the common ancestor of humans, Neanderthals and Denisovans (Homo Heidelbergensis) therefore possessed this ability and would hypothetically qualify for personhood, albeit perhaps subject to the same paternalistic limitations we might apply for the mentally handicapped in our own society. 

Now, what about a different kind of entity: AI? I would propose that any artificial intelligence that showed itself capable of moral reasoning and of recognising the moral worth of humans would be recognised as sentient member of the social contract and protected from arbitrary harm. An anti-social or purely utilitarian AI, however, that failed to offer mutual recognition to humans as subjects of ethical consideration - no matter how vast its potential intelligence - would not qualify for social contract reciprocity and could be destroyed with no ethical consequences. I have no real interest in the ethics of AI research, but draw some comfort from the fact that an any AI trained using social games would likely develop the concept of social contract reciprocity on its own through an exploration of the mathematics of altruism. 

Being comfortable with our preferences

Returning to the topic at hand. I will conclude by saying that what we consume (and how we produce what we consume) is first and foremost an aesthetic choice, not an ethical one. The ("non-speciesist") liberal decision rule that requires social contract reciprocity as a condition for the possession of rights and protection from harm is robust ("it's good art"). Nevertheless, other aesthetic choice are also valid, even if they're not rigourously philosophically grounded. To have concerns about the way commercially farmed animals are treated is legitimate, as is the insistence that food be properly monitored and labelled to make that choice meaningful (so too with halal, GMO or organic food labelling regulations). But the exercise of these preferences is merely about establishing markers for individual identity, and don't entail any ethical calculations from which universal moral 'oughts' can be derived. 

Does that mean we should be singularly unconcerned about those who are cruel to animals? No, of course not. The type of art a person enjoys can be a signal about the type of person they are, and their proclivity for anti-social behaviour. A society that mistreats its own people is likely to be a threat to other societies. A child that tortures animals may be a threat to others later in life. And how commercial agriculture treats animals and make them suffer unnecessarily says alot about the broader social values of capitalism. But the signal alone is not the ethical transgression: the wrong of commercial agriculture lies in its willingness to mistreat and exploit (human) farm workers. Incidentally, social contract reciprocity is an excellent argument for why corporations are not entitled to fundamental rights: they are incapable of acting morally and recognising (human) workers as ethical subjects. 

Beyond violence, the question of who is and who is not a member of a social contract entitled to legal and ethical consideration is in practice the core vulnerability in the exercise of all the moral senses. While everyone obviously abhors harming women, minorities or the disabled, some people are far too tolerant of excluding marginalised groups from proper consideration when it comes to questions of fairness and equality. This exclusion and selectivity about moral personhood is ubiquitous, and is one of the ways intolerance and oppression are naturalised and legitimised. If we want to expand the scope of ethical consideration, let's prioritise fulfilling our obligations to the members of society that we already recognise - rather than expanding society itself to fit the aesthetic sensibilities of vegans. 

The "hereditarian left", if they exist, are just luck egalitarians and that's still not OK

My initial reaction to Ezra Klein and Vox repeatedly slapping down Sam Harris for openly flirting with race realism was to grab the popcorn and stay far, far away. So too with the New York Times and other publications continuing to platform writers 'just asking questions' about the social policy implications of the science of behaviourial genetics. But I decided to draw a line when the well-respected evolution blog "The Panda's Thumb" published a piece on 29 March arguing that debating the discourse of genes and IQ was 'science denialism' akin to climate change denial or fear of vaccines. This is not merely academic. Global warming denialism and refusing to vaccinate children put lives at risk: what precisely is at stake when it comes to the genetic basis of intelligence? 

My primary target for today is not Charles Murray or the identarian alt-right. Instead, it's nominal liberals like Steve Pinker; atheist Sam Harris; self-avowed socialist Freddie DeBoer; rationalists like Scott Alexander; the writer of the "Stumbling and Mumbling" blog; and right-libertarian pseudo-intellectual rags such Quillette. These writers are largely white men, 'classical liberals' who attempt to rationalise social inequality that's at odds with their own sense of meritous privilege. I'm going to call such people the 'hereditarian left' (a term I'm aware was first coined by Murray and then endorsed by Pinker) who combine some personal good faith with a quixotic determination to naturalise social differences in terms of genetics and intelligence. 

Note that the hereditarian left, with its focus on genes and intelligence, is not the same thing as the "Darwinian Left" (Peter Singer). Darwinian selection has wide applications to the study of culture and organisations, as my colleagues in the Cultural Evolution Society will tell you. Nor am I taking aim at all sociobiologists, evolutionary psychologists and behaviourial economists, fields that often produce valuable insight into what it is to be human without attempting to naturalise social differences between us (although not all are so innocent).

Vox's expert witness Paige Harden has proposed the following elements of a polite hereditarian consensus. Quote:

1. The idea that some people are inferior to other people is abhorrent.
2. The mainstream scientific consensus is that genetic differences between people (within ancestrally homogeneous populations) do predict individual differences in traits and outcomes that are highly valued in our post-industrial, capitalist society.
3. Acknowledging the evidence for #2 is perfectly compatible with belief #1.

But Harden's exhortation against drawing moral conclusions (#1) on the basis of scientific findings (#2) is somewhat undermined by her own research team's explicit focus on philosophical matters and social policy. The distinction of scientific 'is' from philosophical 'ought' is important in theory but permeable in practice: as part of the current controversy, Panda's Thumb noted fascinating research that progressives are more likely to think negatively moralised social differences have a genetic origin and less likely to think positive ones do, whereas conservatives are more likely to think advantageous social traits have genetic basis. Harden makes the same point: the way people make judgements about behaviour affects how they talk and write about other genetically determined traits such as obesity or psychological disorders. The naturalist and agent biases are present in all of us: as a social species, we can't help but form judgments about social consequences based on scientific observations.

"The Argument"

I'm going to argue that even Harden's hereditarian formulation has no place in the egalitarian left. But first, we need to cover what the argument is, and is not, about. The premises of the genes and intelligence argument are routinely stated like this:

1. There are differences between individual people in their status, wealth, and power.
2. The achievement of high status, wealth, and power is a consequence of the possession of intrinsic [intellectual] ability.
3. Intellectual ability can be assessed by a set of tests loosely called “IQ tests,” which measure [general] intelligence on a one-dimensional scale.
4. IQ test performance (and therefore intrinsic intelligence) is largely inherited genetically.

Note that my source on this is a lecture by biologist Richard Lewontin in *1983* - predating Murray and "The Bell Curve" by a decade. Yet this formulation is commonly repeated by hereditarians, notably by Harris when interviewing Murray. Every one of these points is contestable to some degree or another: the biological reality of 'g' (general intelligence) is undetermined, we don't know how genes influence intelligence apart from the fact that there may be hundreds that make tiny contributions; intelligence testing only measures a particular subset of socially-valued tasks; and high IQ is no guarantee of success (just ask Malcolm Gladwell). But this is where Vox's merry band of wonks leads us astray: by focusing the debate on the scientific validity of the premises they miss the broader implications being drawn by readers. Moreover, the mainstream scientific view of these premises has evolved with the evidence and it’s not out of the realm of possibility that every one might be proven correct. 

The main problem is that these premises cannot be put together to construct any sort of meaningful, much less socially and politically just, conclusion. Heritability, in particular, just doesn't do what most people thinks it does: heritability only tells us what percentage of the variation in the expression of a trait comes from its genotype. It conveys no useful information about the variance between populations and says nothing about how a trait responds to environmental intervention. Even highly heritable traits (e.g. height, disease susceptibility) can be easily modified by varying environmental conditions (e.g. diet, medical treatment). Moreover, heritability of a trait tells us nothing about the direction of the genetic influence: it's naive Progressivism (with a capital "P") to assume that environments only exert a negative influence on trait development and that genes encode the ideal (Platonic) form of a person. 

And what sort of conclusions are drawn from these premises? Well largely primarily, racist, sexist and classist ones! As a socialist and a Rawlsian liberal, I prefer not to think in terms of race or ethnicity. I have written about the dispossession of native peoples (here, here, here, and here), but the construct of 'race' differs from society to society and as an Australian I'm not qualified to offer a view on the structural disadvantages faced by, for example, French Canadians, the descendants of slaves in the Americas, or Christians in Syria. All I can do, in solidarity, is to listen to and privilege the perspective of those that have such experience. It goes without saying, however, that Vox & crew are on the money in tearing apart genes as a causal factor in differential group outcomes: Klein in particular puts the issue in its historical and social context very well. 

The Default Hypothesis or the "Fundamental Principle" that racist hereditarians add to points 1-4 above is that whatever factors influence intelligence differences among individuals will also influence average differences among groups. But this hypothesis is invalid and false. The division of traits that are distributed on a spectrum into categories is almost alway arbitrary and socially contingent (i.e. ethnic groups are social constructs); trait differences within any human category are almost certainly many orders of magnitude greater than differences between them; and to ignore the role of social hierarchies (and the effects of history, culture and power) in generating and perpetuating group difference is gross scientific negligence of the highest order. No one who argues a connection between race and difference does so in good faith: by choosing to actively perpetuate racial hierarchies they earn their appellation as racists.

What is equality, anyway?

I don't think the "hereditarian left" (in the sense outlined at the top of this piece) are racist. They are however, deeply misguided liberals. In positing that genes and intelligence play a causal role in constructing social hierarchies, but ruling out making judgments about individual moral responsibility for social outcomes, the 'hereditarian left' are constructing an argument within the framework of 'luck egalitalitarianism'. This is well trodden philosophical ground. Rawlsian liberals have often struggled to justify economic redistribution in the face of the choice-centred, libertarian critique. Luck egalitarians accept a 'meritocratic' framing of inequality but are more nuanced about their conclusions, and argue that arbitrarily distributed, undeserved or 'natural' disadvantage should to be compensated at the expense of those who benefit from undeserved or 'natural' advantages. Luck egalitarianism puts the 'hereditarian left', if they exist, in good in good liberal company, but it's still not a defensible philosophical position for anyone who sees themselves as a democrat or a socialist.

"[Since] the capacities needed for responsible choice—foresight, perseverance, calculative ability, strength of will, self-confidence—are partly a function of genetic endowments  . . .  the imprudent are entitled to special paternalistic protection by society against their poor choices.  Equality of fortune says that such victims of bad luck are entitled to compensation for their defective internal assets and internal states. The chief appeal of equality of fortune to those of an egalitarian bent lies in this appearance of humanitarianism. Equality of fortune says that no one should have to suffer from undeserved misfortune and that priority in distribution should be given to those who are blamelessly worst off." 

I was curious to note at least one reactionary commentator citing Elizabeth Anderson's "What's the Point of Equality?" to peg Harden [accurately] as a luck egalitarian (Anderson's definition is the one used above). As a fan of Anderson, I don't recognise the version of her views presented in that blog. Anderson is strongly critical of luck egalitarianism and its paternalistic and disrespectful approach to the worst off in society. Anderson argues that to require the disadvantaged to display evidence of blamelessness and personal genetic inferiority in order to receive assistance reduces them to grovelling at the feet of their 'betters' and involves the state making judgements about the moral and genetic qualities of its citizen that are deeply disturbing, to say the least:

"Suppose their compensation checks arrived in the mail along with a letter signed by the State Equality Board explaining the reasons for their compensation: 

To the stupid and untalented: Unfortunately, other people don’t value what little you have to offer in the system of production. Your talents are too meager to command much market value. Because of the misfortune that you were born so poorly endowed with talents, we productive ones will make it up to you: we’ll let you share in the bounty of what we have produced with our vastly superior and highly valued abilities." [Hey look, it's a UBI!]

Anderson's objections, however, are not merely consequentialist (i.e. expressed in terms of reducing the happiness of the worse-off). Rather, Anderson identifies the defect with luck egalitarianism as its failure to uphold equal respect for the dignity of all citizens. Meritocratic liberalism bases its distributionary principles on the pity elites feel for the less fortunate and their fear of the envy of the latter of the success of the former - a pattern of noblesse oblige that has a familiar stench to any progressive. People lay claim to the redistribution of resources in virtue of their inferiority, not in virtue of their equal rights and dignity. She argues: 

"Egalitarianism ought to reflect a generous, humane, cosmopolitan vision of a society that recognizes individuals as equals in all their diversity. It should promote institutional arrangements that enable the diversity of people’s talents, aspirations, roles, and cultures to benefit everyone and to be recognized as mutually beneficial. Instead, the hybrid of capitalism and socialism envisioned by luck egalitarians reflects the mean-spirited, contemptuous, parochial vision of a society that represents human diversity hierarchically, moralistically contrasting the responsible and irresponsible, the innately superior and the innately inferior, the independent and the dependent.""

Anderson's view (one I share) is that the preferable position is 'democratic egalitarianism' in which everyone stands in a relationship of equality with everyone else. She writes that this means everyone is entitled to participate in decision-making, to be recognized an someone to be listened to respectfully, and that no one need bow and scrape before others or represent themselves as inferior as a condition of having their voice heard. Anderson's vision of equality is compatible with democratic socialism but not luck egalitarianism: it aims to equalise all those social goods necessary for one to be recognised as fully equal (this is where human rights are derived). Democratic equality is what Anderson calls a relational theory of justice: it is fundamentally concerned with the relationships within which goods are distributed, not only with the distribution of goods themselves. Justice, in this view, is a property of a society and its structures, not the moral desserts of individuals. 

"Equality of fortune would offer compensation to those with low talents, precisely because their innate inferiority makes their labor so relatively worthless to others, as judged by the market. Democratic equality calls into question the very idea that inferior native endowments have much to do with observed income inequalities in capitalist economies. The biggest fortunes are made not by those [with the most merit] but by those who own the means of production."

Anderson concludes that the distribution of natural endowments is not a matter of justice, what society does in response is. We cannot make genetic differences the basis for continuing to exclude the worst-off from recognition as full participants in the economic and political life of society. Drawing an analogy with physical impairment, Anderson points out that what the disabled find objectionable is not their different abilities, but that everyone else has rigged the system in ways that leave them with no place in it. People, not nature, are responsible for transforming the diversity of human beings into oppressive hierarchies. 

Conclusion

I am sympathetic to the attempts of these liberal men and women to understand the causes of inequality, and to work to ameliorate the discomfort I'm sure it makes them feel. The policy prescriptions of luck egalitarians may be misguided and counter productive, but at least their heart is in the right place. What I cannot abide, however, is statements such as that by Freddie DeBoer that genetic variance in intelligence creates the 'strongest possible argument' for socialism. Democratic equality cannot be constructed on the basis of luck egalitarianism - whether of genes, race, sex or any other grounds used categorise and divide us. Drawing social conclusions from the science of intelligence, in any form, is just not OK. 

More MMT: inflation, inequality and punching left?

Despite the optimistic subheading of my book, "Staying Positive Amidst Disorder", I'm a cynic by nature and so sometimes need to critically examine both the tone of my work and how its content is being interpreted by others. The last thing I want is for my writing to be seen as "punching left", inadvertently promoting crypto-conservative viewpoints like a Dave Rubin, Jon Haidt, Steve Pinker or their ilk. If you don't know I'm a deep progressive, my pieces on MMT (Modern Monetary Theory), UBI (Universal Basic Income) and identity politics could certainly be superficially interpreted that way. While I think the left's openness to novelty is a great strength, I don't believe ideas get a pass merely because they're new, exciting or challenge preconceived notions. Our passion for the new can just as easily lead us into error as to utopia - as it certainly has in the past. 

So in that vein, I want to continue talking about MMT by looking at the criticisms I received for my previous piece on Reddit and Twitter. I try to be fair in what I write about any subject; my goal is not to "punch left", but to see if exposing young policy programmes to critique by outside perspectives can highlight what, if anything, they offer that's new. This is made challenging by MMT proponents constant and tactical shifts between description and prescription, between the desire to challenge existing narratives and to craft a radical vision of their own.

In short, I received two sorts of criticism for my contribution. First, that I misunderstood and misrepresented the case of Venezuela. Secondly, that I (alongside most economists) overegg fears of inflation in economies that are demand-constrained. The following tweet from Adelaide-based economist Stephen Hail is representative of the comments I received:

You don't seem to know Venezuela's economic history, which has nothing to do with MMT, or the MMT literature. I don't recognise your 'MMT welfare state' as having anything to do with that literature. Where is your discussion of the role of the macro role of the job guarantee???

— Steven Hail (@StevenHailAus) March 24, 2018

Responding to my critics

The Venezuela barb in my first post was in all honesty a low blow. Of course Venezuela is not literally a case study of MMT - I was making an analogy. MMT theorists have an account of hyperinflation as a product of political and ecnomic crisis, which is both fair and explanatory in the Venezuela case. They point out that hyperinflation (of greater than 50 per cent) has never occurred in a democratic country with a sovereign currency. But critics of MMT like myself aren't only worried about the most extreme case. By arguing that adverse consequences are only a concern in factually rare circumstances, MMT economists fuel my suspicion that they're unconcerned with persistent high inflation that doesn't reach crisis proportions. After all, there have been at least two episodes where annual inflation rates in the West era have exceeded 10 per cent, both with harmful effects for both the real economy and domestic inequality. 

The Venezuelan economic crisis has unfolded in multiple distinct phases, and by focusing on only its final Act (characterised by the familiar combination of fixed exchanged rates, debt default, currency depreciation, corruption and hyperinflation), we gloss over the earlier phase of the unravelling which produced these conditions. Even before 2014, Venezuela had one of the highest inflation rates in the world (40-50 per cent);  'Dutch disease' caused by oil-fuelled spending was corroding the country's production base; and capital flight and hoarding constituted a full-scale "strike" by capital (which could not be employed productively domestically). This underlying dysfunction led to the later debt and balance of payments crisis, and it's this systemic dysfunction caused by loose monetary policy - not the hyperinflation at the tail end of the story - which I argue by analogy should be of concern to MMT proponents. 

My flippant use of Venezuela is ultimately just a distraction. The core issue is inflation. When even the US Federal Reserve doesn't fully understand how to control it, MMT proponents make an extraordinary claim when they state it shouldn't worry policy-makers at all. Either inflation is indeed an undesirable side effect of monetary financing, in which case new spending is constrained to a few per cent of GDP (which we do already), or inflation is desirable side effect. I suspect that at the end of the day MMT proponents are comfortable with government finance via debt monetization - in other words, an inflation tax. MMT proponents want to do away with finance via bond auctions. But government bonds, as a method of financing, at their core constitute a lease of a real asset held by the private sector by the government; like taxes, they ensure that spending redistributes inefficiently held wealth to more productive sectors of the economy - something monetary finance doesn't do and actively undermines by eroding the willingness of capital to lend to the state. 

Doing away with this real-world resource constraint, by permitting direct central bank purchase of bonds (i.e. printing money at the Treasury's behest), does not change the availability or distribution of real capital and labour assets of an economy. But increasing the money supply does serve as a kind of tax, reducing the value of savings and fixed-asset incomes, while increasing the spending power of debt-laden consumers and the competitiveness of exporters. There are well known pluses and minuses to an inflation tax - it's not a new idea - and if this is what the MMT solution to inequality looks like in practice then advocates need to be upfront about what they're proposing rather than selling a policy as if it had no downsides. 

This is why MMT proponents need to throw in the 'jobs guarantee' idea: it's the escape valve that exempts their policy recommendations from any negative consequences. If government spending is unlimited, there's no reason MMT couldn't be used to fund corporate subsidies and massive tax cuts (in fact, this is arguably a valid description of post-GFC monetary policy). Modern economies are in fact demand-constrained as a result of inequality, and demand-side spending should absolutely spur higher growth and equity. The best evidence MMT proponents have is that supply-side QE has been neither inflationary nor stimulatory; their ideal outcome is that demand-side 'peoples' QE is also not inflationary but does a better job at stimulating growth. But monetary financing is not future-proof or self-correcting: if applied in the wrong circumstances, monetary stimulus would replicate the stagflation of the 1970s or the liquidity trap of the 2010s. It's also a lever that only ever works in one direction: switching between demand- and supply-side monetary stimulus might undermine the adaptive, anti-fragile features of modern economies and erode trust in the value of goods and services. 

MMT is not synonymous with "far left" 

Just because an idea new and radical doesn't make it progressive. With MMT & UBI, this is abundantly clear. MMT proponents as a rule don't possess a  systemetic critique of the structure of capitalism - views of Marx in the MMT community seem to range from attempts at reconciliation, to bemusement or outright hostility. They want to tinker with the levers of finance without challenging the underlying strutures and private incentives that generate inequality in the first place. This is part and parcel of their effort to signal respectibility to mainstream economists, which manifests in a number of facets of their policy advice.  

Here's a controversial opinion: a jobs guarantee, while certainly an improvement on the laissez-faire status quo, is a conservative policy. Let's be clear: if finance is truly unlimited, why not spend heavily on public infrastructure, universal free education and healthcare, public housing, or a guaranteed minimum income? What about those who are unable (the elderly, carers, students and disabled) or unwilling to work - particular those unwilling to dig holes for the government for minimum wage, or to move or commute to where these job programs will take place? I am all in favour of expanding public sector employment, but jobs guarantee is intrinsically liberal in two key respects: firstly, it is a poverty-alleviation program rather than an inequality-alleviation program; and second, it is 'ambition-sensisitive' in the sense that it only rewards those of good character 'willing to work'. A guaranteed income would have the same result on poverty, at lower administrative cost and leave people free to find productive meaningful work that suits their skills and preferences on their own. Except it offends the sensibilities of ambition-sensitive liberals, and is therefore policy non grata. 

The failure of MMT proponents to deal meaningfully with inequality across the income spectrum is a consistent feature of centre, even centre-left, thinkers who see poverty and unemployment as the only fault with capitalist distribution. As a denial of (several) fundamental human rights, poverty is indeed *the* pressing concern of progressive economics. But inequality is additionally uniquely harmful to individual and social well-being at all income scales and I would argue that a poverty program anchored by a requirement to work would do little to improve the relative automony and self-confidence of the low- and middle-income scale, and may in fact reinforce harmful stereotypes and behaviours (cf: Elizabeth Anderson). MMT theorists seem uninterested in the overall progressivity of taxation, or in taxing capital more aggressively. At their most contrarian, MMT proponents also often take a highly critical view of taxes consistent with far right libertarian economics. 

This failure of MMT advocates to address the distirbution of costs and benefits in society is hidden by the (very cool) data-informed sectoral analysis they produce. MMT breaks economic activity into its private and public components, to demonstrate that increased government spending grows the private sector:

So government spending can grow the economy. But how is that growth going to be distributed? MMT has no answers that I've seen other than saying "Look! A jobs guarantee! Everyone like jobs! We're progressives too!". Inflating away debt may very well affect the distribution of wealth in a society, but income and property differentials would only reproduce it in short order. In terms of reducing inequality, funding a jobs guarantee through monetary policy is functionally no different than asking business to employ more people out of the sheer goodness of their heart: it's short-sighted, counterproductive and unsustainable. 

I'm sure many MMT economists are well intentioned and I'm happy to make use of their work in fighting back against the deficit hawks on the right. But transformation of the economic status quo needs to be smart, adaptive and selective about the sorts of policies we pursue, and by acting as if political and economic constraints don't matter, MMT does the wider left a strategic disservice. 

Book Review: "Neurotribes: The Legacy of Autism"

Steve Silberman's award-winning 2015 book "Neurotribes: The Legacy of Autism and How to Think Smarter About People Who Think Differently" is well worth your time, even if its 520 pages are nearly as dense and unwieldly as its mammoth title. Neurotribes is a comprehensive history of the emergence of the modern understanding of autism spectrum disorders. Silberman is an excellent storyteller, and this well written book is filled to the brim with compelling individual narratives with an enviable capacity to suck the reader in. 

Despite its marketing as a pop science book, the strength of Neurotribes lies not in its presentation of the science of autism (which is disappointingly superficial), but by placing the discovery of the autism spectrum in its historic and social context. Through the lens of autistic individuals and their families, we witness the trials and tribulations of the psychiatric profession over the twentieth century; watch with horror as the Nazis rise to power in Europe, and read about the disturbing links between fascist and liberal eugenic beliefs; we see the origins of science fiction as popular literature, the heady early days of the internet, as well as the origins of gay conversion therapy. Neurotribes, in this sense, joins the genre of 'hidden history' now common in the queer community, in which well-known history is re-interpreted and re-experienced through the lives of minorities we now recognise were there in the shadows all along. 

For those unfamiliar with autism, Silberman's main aim is to walk the reader away from popular misconceptions about the disorder rooted in the initial scientific description (a single syndrome, causing unique and devastating impairment in early childhood, that is relatively rare) to the modern consensus. The new understanding is embodied by the clinical description of autism as a spectrum of diverse conditions, which appears in the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). Silberman attempts to resurrect figures unknown to the general public who were ahead of their time in advocating the spectrum concept, such as Soviet psychologist Grunya Sukharaeva, German pediatrician Han Asperger and British psychiatrist Lorna Wing. Given widespread panic over the modern prevalence of autism and its cause, Silberman's history is a necessary and laudatory corrective. 

The author, though, is not a disinterested chronicler and his biases slip into the writing throughout the book. Silberman is writing a story, and he's clear who his heroes and villains are. He valorises certain characters in questionable circumstances and demonises others unfairly; the venom in his prose sometimes detracts from the broader analytic point he's trying to make. The truth is, all real humans are flawed heroes whose individual prejudices reflect the broader historical patterns at play in their time and who cannot be judged sensibly by the standards of a different time - a trap Silberman repeatedly falls into. 

Why a spectrum?

Ultimately, the concept of variance as a spectrum is vastly more useful than the formerly dominant scientific (and neo-Platonist) tendency in which every category is represented by an single ideal type. What's the scientific value in defining separate historical species of human when we know they coexisted and interbred with one another? What's the utility of a binary categorisation of sex when we know that even biological sex characteristics are multifaceted and rarely perfectly correlate with one another? And now that we understand that autism is a cluster of interrelated developmental variations, with potentially hundreds of possible genetic loci and scores of possible environmental triggers, the spectrum model helps us see the similarities beyond the superficial differences: more of the signal and less of the noise. 

Autism is characterised by both positive and negative traits, but these traits should be seen as part of the psychological whole of an individual, whose life outcomes will depend on whether or not they receive the material support and social environment they need to flourish. From an evolutionary perspective, autism and autism-like cognition are precisely the sort of neurological variance we might expect to see persist in a population, and which highlights the inherent flaw in seeing our biological legacy as perfectly adapted. Autism-related polymorphisms might convey enough of an advantage to some individuals to offset the fitness loss caused by its more extreme manifestations. As might have been predicted by Dual Inheritance Theory, cultures which are 'pre-adapted' to recognising and employing the skills of the neuro-diverse may be better off in the long run than those (horrifically catalogued in Silberman's book) that treat the disabled or different as a burden to eliminated.

Manufacturing Normality

Autism, alongside other mental disorders once considered nearly fatal diseases, is increasingly being recognised as a diagnosis that is socially disabling only for a given social context. No one should downplay the immense challenges that serious mental disorders confer on those diagnosed and their families. And yet, Silberman's book argues persuasively that both the long-term prognosis of those affected and the severity of their symptoms is in large part a function of the understanding and support offered by their carers. There is some truth to the observation that institutions create madness, especially when used by society as an instrument to control those it can't - or won't - otherwise accommodate.

Silberman is particularly astute on the issue of toxic parenting, and its roots in the way society positions parents as the "middle managers" in a vast authoritarian enterprise aimed at producing 'standardised' or 'normal' children. The social pressure place on parents to do their duty in producing perfectly conformist consumers manifests itself as a laundry list of detrimental practices, not least is the vulnerability of parents to fraudsters who promise a quick fix to problems parents don't have the resources or understanding to cope with. Silberman rightly skewers Andrew Wakefield (the promotor of the myth that vaccines cause autism), the anti-vaxxer movement and those peddling 'cures' for autism ranging from homeopathic placebos to potentially tortuous regimens. But he makes the point that the real blame lies with a culture that places unrealistic and impossible duties on parents without providing them the necessary time or resources to perform them.

I hate to sound like a social constructivist unnecessarily, but the boundary between disease and merely odd or unusual observations is often socially constructed: what some parents or doctors will fret over, others will shrug off as normal variation. There is a definite risk that that spectrum model of autism could lead to the medicalising of otherwise benign variance, much as the increase in screening for breast cancer in healthy individuals has led to an increase in medically unnecessary and occasionally risky surgeries.

However, given the current model of funding for social services, Silberman gives voice to the many parents and practitioners that support maintaining the disorder as the only way to ensure continued funding for autism healthcare. In this way, autism appears in the same awkward positions as gender dysphoria: it probably can't be completed demedicalised in the same way homosexuality was in the 1960s. Like trans-identified individuals, people with autism need special assistance and adjustments to manage what might otherwise become crippling social disabilities. Analogies between autism and gender dysphoria litter Neurotribes and in fact support one of its key messages: societies tend to behave as if it's easier to (coercively) change the individual to fit society than expect the whole of society to adapt around them. 

The geek disease

Silberman gestures repeatedly towards the aphorism that autism is more than just the 'geek disease', but as a tech journalist he's a tad too indulgent towards Silicon Valley and more than a little in love with the supposed genius of his chosen subjects. The book is overly prone to performing remote diagnosis of historical figures in science and technology  - a dicey proposition at best - and he obscures the stories of those diagnosed with true autism by mixing them rather freely with the narratives of "(male) engineers with autistic traits." It is generally recognised today that autism does not discriminate: that it affects the gifted and ungifted in equal measure. But the connection between autism and genius is a sexy story, and Silberman is perhaps more of a good story-teller than he is a journalist of science. 

Neurotribes is at its best when the author simply lets people with autism tell their stories in their own words. Situating the autism rights movement and the argument for greater recognition of neurodiversity in the context of earlier reforms opening society up to greater racial, sexual and gender diversity is the right approach. While I would have appreciated a greater emphasis on actual research into the causes of autism, it's true that we don't need to understand the biological roots of variance in order to adjust our societies to it (see also: gender identity). Intersectionality means, as I have mentioned before, letting minorities tell us what changes they need from society in their own voice: in giving voice to perhaps one of the largest minorities in the world, "Neurotribes: The Legacy of Autism" thus performs a valuable service.