F. H. Bradley was better than the caricature Russell needed

Every discipline keeps a founding story, and the one analytic philosophy tells about itself has a villain. In My Philosophical Development Bertrand Russell dated his liberation precisely: "It was towards the end of 1898 that Moore and I rebelled against both Kant and Hegel." He remembered the feeling as "a great liberation, as if I had escaped from a hot house onto a windswept headland." The hot house had a proprietor. Kant and Hegel were dead; the man Russell and Moore were actually arguing with was alive, in Oxford, a fellow of Merton College who had published the most formidable metaphysical system in English, and who would go on answering them in Mind. Francis Herbert Bradley died on 18 September 1924, three and a half months after King George V made him the first philosopher to receive the Order of Merit. The rebellion he provoked has been running the profession ever since.

The story as usually told has the rebels winning by argument. Bradley held that relations were unreal and reality a single seamless Absolute; Russell showed that mathematics requires real relations, Moore showed that propositions are not mental, and the fog lifted. It is a good story, and it makes a discipline's founding look like a proof. The case made here is that it is wrong in two places. The regress argument about relations in the third chapter of Appearance and Reality (1893) has never been refuted; it has been declined, ignored and worked around, which is a different thing. And the moral philosophy of Ethical Studies (1876), the book the caricature forgets he wrote, set out in the 1870s the objection to liberal individualism that communitarians would rediscover in the 1980s. Bradley was a problem the twentieth century postponed, not a mist it cleared.

A fellowship with no duties, and a man who used it#

Bradley was born in Clapham on 30 January 1846, a son of Charles Bradley, a prominent Evangelical preacher. He went up to University College, Oxford, in 1865 as a scholar, took a first in classical moderations in 1867 and only a second in Greats in 1869. In December 1870 Merton elected him to a fellowship of a kind that no longer exists: tenable for life, with no teaching duties, terminable only on marriage. He never married, and he held it until his death, nearly fifty-four years later. In June 1871 he suffered a severe inflammation of the kidneys with permanent effects, and thereafter lived as an invalid and a recluse. R. G. Collingwood, who would inherit much of his outlook, recorded in his Autobiography that "although I lived within a few hundred yards of him for sixteen years, I never to my knowledge set eyes on him." The one biographical detail everyone remembers is that he liked guns and disliked cats, and reconciled the two preferences by shooting the latter in the college grounds at night.

Merton's fellowship was a machine for producing books, and it produced four. Ethical Studies (1876) attacked utilitarianism and Kantianism together and offered a Hegelian ethics in their place. The Principles of Logic (1883), published a year before Frege's Grundlagen, argued against the psychologism of Mill that the content of a judgement is not a mental image but a universal meaning, a thesis Russell absorbed before he turned on its author. Appearance and Reality (1893) was the system; Essays on Truth and Reality (1914) its defence. The preface to Appearance and Reality contains the sentence that will outlive the rest: "Metaphysics is the finding of bad reasons for what we believe upon instinct, but to find these reasons is no less an instinct."

The rebellion was against a person#

The idealism Russell abandoned was Bradley's as much as Hegel's, and the break was a break with a named and living opponent. The doctrine at issue was internal relations: the view that a thing's relations to other things belong to its nature, so that nothing can be fully understood in isolation and, at the limit, no statement about an isolated object is true without qualification. Russell saw that his logicist programme needed asymmetrical relations such as "greater than" to be perfectly real and perfectly external to their terms. If Bradley was right about relations, there was no arithmetic. Moore's "The Nature of Judgment", in Mind in 1899, cut the other cord, insisting that the constituents of a proposition exist independently of any mind that thinks them. Russell's "On Denoting" (1905) carried the campaign into the jokes, with the present King of France, who is neither bald nor not bald: "Hegelians, who love a synthesis, will probably conclude that he wears a wig."

Bradley did not go quietly. He and Russell fought an extended engagement in Mind in 1910 and 1911, Russell offering "Some Explanations in Reply to Mr. Bradley" and Bradley returning with a "Reply to Mr. Russell's Explanations", in the tone of two men who had each concluded that the other was refusing to see the point. The best recent account of the quarrel is Stewart Candlish's The Russell/Bradley Dispute and its Significance for Twentieth-Century Philosophy (2007), which argues that Bradley's position on relations was "unanswerably correct", and that Russell's victory was won on temperament and institutional succession rather than logic. James Levine, reviewing the book for Notre Dame Philosophical Reviews, disputed a good deal of it and still conceded the central claim: that "the victory of Russell over Bradley, at least as a matter of historical fact, was not determined by the merits of the arguments." That is an unusual concession from a historian of analytic philosophy, and it is the hinge of what follows.

The regress was bypassed, not answered#

Here is the argument as Bradley gave it in chapter III of Appearance and Reality. Take two terms, A and B, and a relation R that is supposed to hold between them. If R is something real over and above A and B, then we may ask what connects R to A and R to B. Whatever we answer will be a further relation, which will require a further connection in turn. "The links are united by a link," Bradley wrote, "and this bond of union is a link which also has two ends; and these require each a fresh link to connect them with the old." If, on the other hand, R is nothing over and above its terms, then it does no relating at all, and A and B are left exactly as unconnected as before. Either way, relational thought cannot deliver a unified fact, and Bradley's conclusion was that relations are appearance: indispensable to thinking, incapable of being finally real.

Russell's answer, given in various forms, was that it is the job of relations to relate; a relating relation is a primitive that requires no further relation to do its work, and to ask for one is to misunderstand what a relation is. Frege's apparatus offered a different escape: concepts are "unsaturated", incomplete entities that need completion by objects, so that constituents fit together without intermediaries. D. M. Armstrong posited a "non-relational tie" of instantiation, a fundamental nexus that is explicitly not a relation and therefore not subject to the regress. Wittgenstein's Tractatus has objects hanging in facts "like the links of a chain", with no further glue.

None of these is a refutation. Each accepts Bradley's premise, that ordinary relations cannot unify their terms, and then introduces a special item that can, on the ground that it is not the kind of thing the argument applies to. The Stanford entry on Bradley's regress, surveying the literature, notes that most of those who take the argument seriously find that the proposed solutions "either face their own regress problems or shift rather than solve the underlying puzzle about how constituents achieve genuine unity." A non-relational tie is a relation that has been asked to sign a document promising not to be one. Michael Della Rocca, in a 2024 essay, put the point more sharply: Russell's actual reply to Bradley was that "the search for a 'sufficient reason' is mistaken", which is to say that Russell rejected the demand for an explanation of unity rather than supplying one, and the tradition followed him without noticing that a denial is not an argument. Della Rocca is a partisan, and one should say so; his own The Parmenidean Ascent (2020) takes the regress all the way to its monist conclusion, and few metaphysicians will go with him. But the question of how a fact is more than a list of its constituents remains open in the philosophy of truthmakers, of states of affairs and of grounding, under names that hide its ancestry.

There is a reasonable reply, and it deserves its paragraph. The rejectionist, as the Stanford entry calls him, says that Bradley simply assumes relations cannot relate without a further ground, which is exactly what a realist about relations denies; the regress is a demand for explanation where none is owed, and the correct response is to decline it. That may be right. But declining a demand is a stance, not a proof, and the founding story is told as a proof. Analytic philosophy began by changing the question, from "how is unity possible?" to "what follows if we take it as given?", and the change was enormously fruitful and never justified by the standards the movement set for everyone else.

The Absolute, and the poet it reached#

The second half of Appearance and Reality is harder to defend and less often attacked. Having argued that every category of ordinary and scientific thought, from primary and secondary qualities through space, time, causation and the self, contains contradiction, Bradley concluded that reality must be one, must be experience, and must be a whole in which all appearances are harmonised. Almost nothing can be said of the Absolute beyond its consistency and sentience. This is where the caricature gets its material, and the preface line about bad reasons and instinct was written by a man who knew how his conclusions would sound.

It was also the part that travelled. In 1916 T. S. Eliot completed a Harvard doctoral dissertation, "Knowledge and Experience in the Philosophy of F. H. Bradley"; Josiah Royce praised it, and Eliot, by then in London, never returned for the viva. Faber published it in 1964, with a preface in which its author called the work a curiosity of biographical interest. Eliot's note to line 411 of The Waste Land quotes Appearance and Reality, page 346, on the privacy of experience: "My external sensations are no less private to myself than are my thoughts or my feelings. In either case my experience falls within my own circle, a circle closed on the outside." The sealed selves of that poem, each in its prison, thinking of the key, are Bradley's finite centres of experience with the theology removed.

My station and its duties, a century early#

The book the caricature forgets is the first one. Ethical Studies appeared in 1876, two years after Sidgwick's Methods of Ethics, and it reads like a demolition conducted by someone who has decided that both of the century's ethical theories are empty rather than false. The third essay, "Pleasure for Pleasure's Sake", argues that hedonism sets an end that cannot be reached: pleasures are a perishing series, each gone once felt, so the "sum" of pleasure utilitarianism tells us to maximise is not a thing anyone could possess. The fourth, "Duty for Duty's Sake", finds the categorical imperative formally correct and materially blank; it tells you to will universally, and not what to will.

Essay five, "My Station and Its Duties", ought to have made him a permanent reference in political theory. Its claim is that the self Mill and Kant both start from, the individual prior to society choosing his ends, is a fiction. "The 'individual' apart from the community is an abstraction. It is not anything real." The child, Bradley writes, learns to speak and "appropriates the common heritage of his race"; "the soul within him is saturated, is filled, is qualified by, it has assimilated, has got its substance, has built itself up from, it is one and the same life with the universal life." Morality, on this view, is the realisation of an already social self in the roles and obligations its community assigns. He was blunt about the corollary. "The mere individual is a delusion of theory; and the attempt to realize it in practice is the starvation and mutilation of human nature." And, in a sentence that has made liberal readers wince for a hundred and fifty years, "to wish to be better than the world is to be already on the threshold of immorality."

Set that beside Alasdair MacIntyre's After Virtue (1981), Michael Sandel's Liberalism and the Limits of Justice (1982) or Charles Taylor's Sources of the Self (1989), and the resemblance is not loose. The "unencumbered self" Sandel found in Rawls is Bradley's "mere individual"; MacIntyre's narrative self, intelligible only through the roles and traditions it inherits, is a station with its duties. The communitarians reached these positions largely through Aristotle and Hegel, and the Stanford entries on Bradley's ethics do not claim him as their ancestor. The point is one of priority, not influence. The objection they raised against Rawlsian liberalism had been raised, in English, in Oxford, against Mill, a century earlier, and forgotten because the man who raised it was on the losing side of a different argument.

It is also worth saying, because the caricature will not, that Bradley did not stop at the station. The essay's closing pages concede that the social self "has not succeeded" in giving a complete account of morality, that the satisfaction it offers is "neither ultimate, nor all-inclusive, nor anything but precarious", and the sixth essay, "Ideal Morality", goes past one's community to standards it may not meet. Sidgwick, reviewing the book in 1876, identified Bradley with "my station and its duties" and stopped there; most readers since have done the same. Bradley anticipated the communitarian turn and its liberal rejoinder in one volume, which is more than either side managed in the 1980s.

Why he was hard to keep#

In the Birthday Honours gazetted on 3 June 1924 he received the Order of Merit, an honour then in its third decade that had never before gone to a philosopher. He died of blood poisoning in Oxford on 18 September and is buried in Holywell cemetery, a few hundred yards from the college he barely left; This Day for 18 September records it. Within a generation the profession had decided that his death closed an era, and Russell's memoirs, written with a wit Bradley never attempted, fixed the version that survived.

What was lost in that version was not a system anyone is likely to revive; the Absolute is gone. What was lost was a standard. Bradley asked how a fact is more than a list, how a self is more than a chooser, and why a theory of the good should be believed if its end can never be possessed, in a prose that gave no quarter and expected none. Analytic philosophy answered the first question by changing the subject, the second a century late under other names, and the third not at all. Better than the caricature is a low bar. He clears it by the height of the chapter on relations, and the profession that defined itself against him has not yet said what is wrong with it.