Why you should read Chinese philosophy: Mozi

My aim here is to present the work of Mozi, a classical Chinese philosopher comparatively underknown in the West. In my view, he is one of most interesting philosophers in all traditions, and in telling his story I hope also to make the case for the interestingness of Chinese philosophy.

Once I was talking to Hong Kong students about classical Chinese philosophy. They were dismissive: it’s not real philosophy, real being the sort of stuff called ‘analytic’, they said.

It’s not hard to have that impression. Confucius is the perhaps the biggest name, but if you’ve ever skimmed his work, or even encountered it in pop culture, you may get the case that it’s just a bunch of gnomic and unargued for assertions, things like — to pick some examples at random:

• The noble person seeks harmony but not sameness; the petty person seeks sameness but not harmony.
• To make a mistake and not correct it — that is the real mistake.
• When I walk with two others, there is always something I can learn from them.

Similarly, you may well have encountered the Dao De Jing which equally is heavy on the gnomic and unargued-forness, perhaps most famously its confusing opening sentence:

• The Dao that can be followed is not a constant Dao; the name that can be named is not a constant name

Faced with these you might think that it’s essential that philosophy have arguments for the views it presents, these seem not to, so they aren’t philosophy. Without pointlessly trying to define ‘philosophy’, this is a somewhat reasonable thing to say. But at the same time, always relying on an argument risks, to use a term of art, undergenerating: of classifying some examples of philosophy as not philosophy. Thus the preSocratics telling us that everything is water or whatever don’t really argue for it. Vast swathes of Wittgenstein’s Tractatus are just assertions. Spinoza says he has arguments but they’re not great.

Something can be philosophically valuable even if not argumentatively convincing, as showing a new picture of how a bit of reality might be; and that’s a reason to attend, perhaps, to Confucius and Laozi.

But even if you’re not convinced by that sort of thought, other Chinese philosophers do have arguments. Xunzi, Zhuangzi (ish), and definitely Mozi (he, as we’ll see, argues almost pathologically: argues even when he doesn’t have anything good to say)

Before looking at what Mozi claims, I first want to briefly say a bit about Confucius. He was already when Mozi was writing a central figure, and features, as an argumentative foil, in all the authors just mentioned. His asserted-unargued view, I reckon, is kind of like retvrn for 5th century BC China. Retvrn guys find in the past their idea of the good life, and view the problems of the present as resulting from the decline away from the good old days, from an era of the nuclear family, early marriage, religion, unprocessed food, paintings that look like the thing they’re supposed to look like, and so on. Confucius too thinks things aren’t as they used to be, that old texts are to be revered and learned from, that old habits (as set out in one of the founding texts of Chinese civilization, the Book of Rites) are to be followed, and that, most importantly, people know their role: that fathers act as fathers, sons as sons, rulers and rulers, and so on. Everyone has a place in society, and society is good if people adequately inhabit their place.

Mozi comes about a century after Confucius; the latter was sufficiently important as to warrant a couple of chapters in the Mozi, which represents a collection of writings by which we know the views of Mozi and his followers (which we call Mohists). With that said, let’s get to it.

Mozi’s range is impressive in a couple of ways. He covers a bunch of issues in descriptive (saying how something is) and normative (how it should be) political and moral theory, with important parts about applied issues, such as war and economics. There is stuff (albeit not in my view interesting) about epistemology and methodology. And also ghosts!

It proposes conclusions iconoclastic for the society of its time, challenging deeply held views and entrenched practices. And it does so, at least some of the time, by using, many centuries before anyone else, consequentialist reasoning.

If Confucius is a retvrner, Mozi is well-read, in several different ways, as an effective altruist. In both groups we see counterintuitive views presented boldly by going where the moral mathematics leads. The impressive thing is that Mozi was doing this 2,500 years ago, an intellectual breakthrough that seems more incredible the more I think about it.

There’s another modern type Mozi could be mapped onto: the troll. I say this because, at times, the sophisticated arguments drop out and are replaced with utterly terrible ones. One gets the sense this is someone who wants to get some conclusions accepted and will use whatever arguments will convince, something that while not particularly intellectually honest does make the book entertaining.

(The more sober conclusion, that different parts are written by different people, is almost certainly true, but let me trollishly not let that fact get in the way of my enjoyment.)

Let’s start with the consequentialism in action. By consequentialism I mean a moral theory that considers an action right to the extent it produces good outcomes. Obviously we should say what ‘good’ here means; people disagree, but one simple way would be to think in terms of happiness or pleasure. An action is right to the extent it makes people happy; one action is better than another if it leads to more happiness. The good thing must be in a sense quantitative; the goodness of an outcome can be compared or summed with the good of another; we should seek for the outcome containing maximum goodness, which we can determine by — say — summing the happiness or pleasure or whatever of the people in the outcome.

Mozi lived in a society with a bunch of war (that era in Chinese history is known as the ‘warring states period’), with a bunch of separated areas presided over by leaders who lived in luxury. Much of his best work is a direct response to that. So, firstly, he’s a pacifist for simple consequentialist reasons. Here’s a relevant passage:

The murder of one person is called unrighteous and incurs one death penalty. Following this argument, the murder of ten persons will be ten times as unrighteous and there should be ten death penalties; the murder of a hundred persons will be a hundred times as unrighteous and there should be a hundred death penalties. All the gentlemen of the world know that they should condemn these things, calling them unrighteous. But when it comes to the great unrighteousness of attacking states, they do not know that they should condemn it. On the contrary, they applaud it, calling it righteous. And they are really ignorant of its being unrighteous. Hence they have recorded their judgment to bequeath to their posterity. If they did know that it is unrighteous, then why would they record their false judgment to bequeath to posterity? Now, if there were a man who, upon seeing a little blackness, should say it is black, but, upon seeing much, should say it is white; then we should think he could not tell the difference between black and white. If, upon tasting a little bitterness one should say it is bitter, but, upon tasting much, should say it is sweet; then we should think he could not tell the difference between bitter and sweet. Now, when a little wrong is committed people know that they should condemn it, but when such a great wrong as attacking a state is committed people do not know that they should condemn it. On the contrary, it is applauded, called righteous. Can this be said to be knowing the difference between the righteous and the unrighteous? Hence we know the gentlemen of the world are confused about the difference between righteousness and unrighteousness.

We all know murder is wrong, as does Mozi. But notice he immediately, and I would say unexpectedly, makes the move to quantify the badness, in terms of punishment. I suspect one just doesn’t see that sort of move in Greek or Indian texts.

(In Plato, you definitely get the idea that there are degrees of rightness or wrongness, and that we need some abstract Right to sort of ground these different rights or wrongs, but the further move to viewing them as on a scale isn’t there. Correct me if I’m wrong! And some Buddhist texts certainly have numbers, like “84,000 billion nayutas [a big number] of buddhas”, but it’s for rhetorical effect).

More generally, something we know from the history of science and ideas in general that the idea of measuring something is an idea difficult to come by. Aristotle was no slouch, but famously we had to wait for Galileo to check his theory of gravity by actually measuring how things fall. It has been argued that the idea of probability came very late — that before long into modernity, while we knew that one person might be more likely to die sooner than another, the idea that they were x times more likely to die in 10 years was hard won, and lucrative when won.

So I think this is very impressive, millennia before Bentham, and moreover seems like a pretty solid argument against offensive war. In addition, I’d just note how rhetorically impressive it is as a single tight paragraph of reasoning. You might think this isn’t really deserving praise, but I think it compares favourably to basically any ancient from any tradition. (Try reading Aristotle, you can spend ages trying to work out exactly what the argument is.)

Before going on to consider an objection that might have occurred to you, especially if you’re familiar with debates about consequentialism, let’s see this sort of reasoning in action elsewhere.

Mozi is against a lot of things: music (ceremony in general); lavish funerals and mourning periods; fancy clothes or food or housing. You might imagine a hearer of Mozi’s anti-war argument complaining that without pillaging others their living standards will worsen. But he has a reply. If clever people rule, the benefit (利, the thing maximized in the consequentialist calculations) will increase. And:

This increase is not by appropriating land from without. But by cutting out the useless expenditures it is accomplished. In issuing an order, taking up an enterprise, or employing the people and expending wealth, the sage never does anything without some useful purpose. Therefore wealth is not wasted and people’s resources are not exhausted, and many are the blessings procured. In making clothing, what is the purpose? It is to keep out the cold in winter and heat in summer. The good of clothing is measured in terms of the amount of warmth it adds in winter and coolness in summer: what is merely decorative and does not contribute to these is to be let alone. In building palaces and houses, what is the purpose? It is to keep out the wind and the cold in winter and heat and rain in summer, and to fortify against thieves. What is merely decorative and contributes nothing to these should be let alone.

So: if one wants to maximise benefit, just ensure that your clothes etc. do their job and stop there. Similarly, don’t spend tons of money on funerals or go into mourning for years; that’ll just lead to less beneficial outcomes.

Finally, in a similarish vein, he even has something to say about a 2,500 bc Chinese loneliness epidemic. Like in modern China, there are too few women for single men. The problem back then wasn’t the one child policy but the fact that the local kings took large harems:

The present rulers of large states retain as many women as a thousand in their household and those of small states as many as a hundred. Therefore men in the empire are mostly without wife and women without husband. The functions of men and women are prevented and the population becomes small. If the rulers sincerely desire the population to be large and hate to see it small, they must not indulge in retaining too many women.

(Obviously you can bridle at the thought that women are to be treated as akin to property that can be rightly or wrongly distributed; we don’t see any proto-feminism in Mozi, but at least nothing going unattractively in the other direction.)

In addition, just like Elon Musk, Mozi has concerns about the birth rate. But unlike Musk, he recognizes that it’s the unfavourable conditions in which people are living in, as a result of their rulers’ actions, that lead to this. Having noted that more people are having kids later, decreasing the average number of kids, he proposes an explanation:

Those who govern the empire to-day diminish the people in more ways than one: Employing the people they exhaust them, levying taxes they make them heavy. People fall into poverty and innumerable persons die of hunger and cold. Moreover the rulers make war and attack some neighbouring states. It may last a whole year, or, at the shortest, several months. Thus man and woman cannot see each other for a long time. Is not this a way to diminish the people? Living in danger, eating and drinking irregularly many become sick and die. Hiding in ambush, setting fire, besieging a city, and battling in the open fields, innumerable men die. Are not ways of diminishing the people getting numerous with the government of the rulers of to-day?

Just as an aside, it’s wild that if you judiciously replace various words with ‘late capitalism’, ‘bullshit email jobs’ you could get a complaint people today face. But note how his complaint dovetails neatly with what he’s said before: it’s unbeneficial wars and unnecessary expenses that lead to these problems. If we follow his advice and do our moral mathematics properly, we’ll solve all sorts of societal problems. His is a view in which everything is worth thinking about, arguing about, fixable. Repeating myself, I don’t think you can find any other ancient philosophy with simultaneously such a broad range of topics and such a unified framework for viewing them.

But let’s get back to consequentialism. Although I don’t think we quite saw it in the quoted passage, you might think that for this line of reasoning to work we’re going to have to need a claim to the effect that killing, and more generally doing bad, is always equally bad wherever it happens. If I harm the person from the opposing state, it’s as bad as if I harm my neighbour.

You might deny this. If you’re familiar with consequentialism, you’ll know that there’s an objection in the vicinity. For standard consequentialism, a life is a life is a life. The death of someone far away is as bad as a death around the corner; that equality enables us to do the math. And that equality drives EA’s concern for, for example, mosquito nets: limited as I time in time, space, and resources, I should concentrate on doing the most good I can do. Saving lives is good; but all lives are equally worth saving, and some lives are easily savable, or at least substantially improvable. So, I should buy mosquito nets for citizens of a poor country that can’t afford them where mosquitos are lead to easily preventable deaths.

But, the response goes, this is wrong. Some lives matter more to us: those of our loved one, including perhaps especially ourselves. Consequentialism can’t account for this partiality of ethics, so consequentialism is no good.

Note that the Chinese society in which Mozi found himself was one in which partiality runs deep. A central plank of the Confucian view was what is called 孝, xiao, filiality. It was very important that one, for example, respect one’s father. If you can buy a malaria net or chip in for some glp-1’s for your fat dad, the Confucian would do the latter.

So at minimum Mozi at this point owes us something: something going against this idea that ethics should be partial. Thankfully he has something, his view of universal love, or inclusive care. The Chinese phrase here is 兼愛, the former meaning together, jointly, and the latter being the word for love, still used up to today. According to Fraser (The Philosophy of the Mozi, an accessible intro to his work), this word ranges from “strong affection to dispassionate concern”. It can refer to “a parent’s love for a child, fondness for someone’s physical appearance, concern about the health of one’s livestock, or concern for the welfare of one’s political subjects”.

In Mozi’s view, the partiality mentioned above, the partiality that would make consequentialism tricky, leads to big problems.

[When it comes to thieves and robbers] As he loves only his own family and not other families, the thief steals from other families to profit his own family. As he loves only his own person and not others, the robber does violence to others to profit himself. And the reason for all this is want of love. This again is true in the mutual disturbance among the houses of the ministers and the mutual invasions among the states of the feudal lords. As he loves only his own house and not the others, the minister disturbs the other houses to profit his own. As he loves only his own state and not the others, the feudal lord attacks the other states to profit his own. These instances exhaust the confusion (亂) in the world.

So, things like war lead to the moral calculations we saw; and things like war arise from the lack of this inclusive care. And so we should promote this inclusive care. Mozi has at least begun to discharge his argumentative duty.

Of course, more would need to be said. You might immediately wonder: but is this realistic? It would be all well and good if we were like that, but it’s simply a fact that we are partial creatures, and there’s no getting above that. This is what’s called a demandingness objection, and again, Mozi is on it.

He says a few things here. One is that if we actually were to practice this, we would see the benefits, the 利. If you care for someone, they’ll care for you; if you benefit someone, they’ll benefit you.

He also proposes the following argument, which sounds to me reminiscent of Parfittian stuff on (in)direct self-defeatingness:

[Imagine two rulers, one partial, one impartial, they], then, are opposed to each other in word and also in deed. Suppose they are sincere in word and decisive in deed so that their word and deed are made to agree like the two parts of a tally, and that there is no word but what is realized in deed, then let us consider further: Suppose, now, that there is a disastrous pestilence, that most people are in misery and privation, and that many lie dead in ditches (Under such circumstances) let us inquire, if a person could choose one of the two rulers, which would he prefer? It seems to me on such occasions there are no fools in the world. Even if he is a person who objects to universal love, he will choose the universal ruler. This is verbal objection to the principle but actual selection by it — this is self-contradiction between one’s word and deed.

Assessing this is tricky, so let’s not try. We can note that it does seem like he’s making pretty big claims about the (personal) benefit of impartiality which one can definitely query.

But maybe — though at this point I’m getting a bit galaxy-brained — perhaps one can see a response in another thing he says:

It seems to me that the only trouble is that there is no superior who encourages it. If there is a superior who encourages it, promoting it with rewards and commendations, threatening its reverse with punishments, I feel people will tend toward universal love and mutual aid like fire tending upward and water downwards — it will be unpreventable in the world.

The thing is, we’ve missed an important part of the theory. Mozi seems much more comfortable than we do with the idea of there being superiors who can compel behaviour from us. He happens to think that the compelled behaviour (inclusive care) is to the benefit to the compellee; but even if this isn’t so, there’s still a way, albeit one unattractive to us, of bringing about his model of inclusive care.

Before rounding out this conceptual bundle by considering a final piece of Mozi’s theory, I want to make good on my ‘troll’ claim. Hopefully you’re on board with the claim that we have an incredibly sophisticated thinker here, one who can easily go toe-to-toe with any of his axial age fellows, and indeed with many of the best philosophers from any period of history. But at the same time there are claims and arguments in Mozi that are completely terrible.

We see this already here. One of his main argumentative moves is, in essence, an appeal to authority: to the ancient or older rulers of China. They practiced inclusive care, so we should too:

The “Great Declaration” proclaims: “King Wen was like the sun and the moon, shedding glorious and resplendent light in the four quarters as well as over the Western land.” This is to say that the love of King Wen is so wide and universal that it is like the sun and the moon shining upon the world without partiality. Here is universal love on the part of King Wen.

We see this sort of thing again when he argues for the existence of ghosts, where basically the whole thing is an appeal to authority. In some cases the supposed ancient exemplar is better than the one cited, but it is still odd for a person who argued against traditional mourning because it wasted resources (the lavish funeral, the fact the man couldn’t work for years, etc.) to appeal to tradition in this way. It’s basically unbelievable to me that someone otherwise so sophisticated could say such things, and I’m tempted to think that what we’re seeing here is someone attempting to convince his listener of his view by any means. Hence: troll.

(Again, I suppose the sober thing is to say is that we’ve got different authors.)

Digression: classical Chinese.

A further reason to be interested in Chinese philosophy in general is the uniqueness of the classical Chinese language. Classical Chinese is what we might call gigantically semantically undetermined: what a sentence expresses when used goes considerably beyond what the words it contains literally mean.

This is because classical Chinese doesn’t have number, gender, modality; while some languages (not really English) distinguish between a basic form of the verb and a causative version (‘rise’ and ‘raise’ is one of the few examples in English), Chinese doesn’t. A given character can often function as a verb or a noun. Here’s a famous example showing all this, from Confucius:

子曰：「君君、臣臣、父父、子子。

The first two characters refer to Confucius. The second two consist of the word for gentleman twice; the third two, minister twice; then father twice, then son twice.

This should be meaningless! But the second minister (gentleman, etc.) functions as a verb: act like a gentleman. So a final interpretation ends up being:

• Confucius said: ‘the gentleman should act like a gentleman, the minister should act like a minister, the father should act like a father, the son should act like a son’

A consequence of this is that any given sentence has a ton of potential meanings, and a lot, in theory, is on the shoulders of the reader. An odd thing, however, is that very often in fact context does suffice to determine a clear meaning; what one might have thought as as a gigantic problem is in fact not one, which is an inherently interesting fact about how language works.

Hobbes avant la letter

So far, hopefully, so interesting. I want to end by expanding on the idea just floated of having an authoritarian figure to fix society, one who can force us to love one another.

Western philosophers are of course familiar with this move. We see this most famously in Hobbes, whose state of nature is one of the most famous, quoted, and influential bits of Western philosophy. The English-civil-war-enduring Hobbes considered a hypothetical pre-society state in which there was no summum bonum, no highest human good we could all agree upon, which society could decide to unifiedly pursue. In the absence of this, there is no unity, everyone is for themselves, and, in words it’s obligatory to quote:

In such condition there is no place for industry, because the fruit thereof is uncertain, and consequently no culture of the earth, no navigation nor the use of commodities that may be imported by sea, no commodious building, no instruments of moving and removing such things as require much force, no knowledge of the face of the earth, no account of time, no arts, no letters, no society, and which is worst of all, continual fear and danger of violent death, and the life of man, solitary, poor, nasty, brutish, and short.

And he thought that to solve this problem we needed to give up our rights to retaliation, vesting them in the sovereign leviathan, proposing this as a sort of picture of how society in general evolved.

Mozi lived in similarly troubled times, and he has his own version of Hobbes’ state of nature:

In the beginning of human life, when there was yet no law and government, the custom was “everybody according to his own idea.” (人異義) Accordingly each man had his own idea, two men had two different ideas and ten men had ten different ideas — the more people the more different notions. And everybody approved of his own view and disapproved the views of others, and so arose mutual disapproval among men. As a result, father and son and elder and younger brothers became enemies and were estranged from each other, since they were unable to reach any agreement. Everybody worked for the disadvantage of the others with water, fire, and poison. Surplus energy was not spent for mutual aid; surplus goods were allowed to rot without sharing; excellent teachings (Dao) were kept secret and not revealed. The disorder in the (human) world could be compared to that among birds and beasts.

人異義 consists of the character man, distinguish/separate, and right, but we should understood ‘right’ here are really meaning something like the word ‘right’’ or concept of right. So it’s saying that the problem arises from our having different and competing conceptions of what is right.

Note that these are remarkably similar views: both think that in the absence of a shared good we all know each of us is pursuing, and which we can orient society to pursue, we are left in an anarchic position. There is however one interesting difference, arguably. Fraser points out that it seems as if the badness of this original position is different for the two. One way to see this is to consider the final bits of both quotations. For Hobbes, the badness is in the ‘life of man’. But for Mozi, it’s 亂, disorder, confusion, where such things are society-level problems.

The Hobbesian state-of-nature is individualistic anarchy — there is no ruler, there isn’t really a sense that people have come together into society. But for Mozi, the analogous state is still a society, it’s just a bad, disordered society. You might perhaps think that for a Chinese person of the era, that society exists is something that can’t be questioned, which would fit well both with what we know about Chinese thought and also stereotypes concerning the Chinese (as being non-individualistic) that exist to this day. Different broad philosophical ideas — the state of nature — seem to appear in the two very different traditions, but they are in each case tailored to the society they appear in.

But let’s return to the problem: things are going back because we all have different concepts of 義 or right. How do we solve that problem? Mozi’s answer is that we coordinate on one single conception of right, offered by the emperor

What the emperor asserts, everyone should assert; what the emperor rejects, everyone should reject.

天子之所是，皆是之，天子之所非，皆非之

We basically outsource our assessing what is right to one figure; so agreement concerning right is guaranteed. If we do, perhaps along with inclusive care, we’ll see an end to the disorder that plagues society.

So, that’s some Mozi. Hopefully you agree he’s worth your attention. If you find this interesting, here’s a bunch more essays about various aspects and schools of philosophy.