Thanks, whoever moved this here. Particularly if it was me.
Moving forward - almost finished!
Analytic Philosophy began in 1903, with GE Moore’s “Refutation of Idealism” and “Principia Ethica”, helped along by Bertrand Russell’s “Principles of Mathematics”. It’s never been entirely clear what Analytic Philosophy is – indeed, the later stages have entirely abandoned most of what was distinctive about it originally. Rather than being a system or a programme, it should probably be seen as a style or as a communal expedition. However, beginning with Russell and Moore, three principle features arose that marked a new style of philosophy:
1. A concern for such things as ‘meaning’, ‘propositions’, ‘logic’ and ‘language’.
2. An eschewing of grand ‘systematising’ and ‘synthesis’, in favour of the analysis of small, individual problems.
3. An emphasis upon “clarity”, “rigour” and “precision”, rather than rhetorical power or inherent poetry.
4. Not being Hegel. Not content with not actually being Hegel, the early Analytics were determined to be as unlike Hegel as possibly possible – Russell described the project as being driven by the desire to believe everything Hegel had disbelieved, and to disbelieve everything Hegel had believed. This was not a hatred born of ignorance, but of experience – Russell, like the other early Analytics, began as ardent Hegelians in their own right, before feeling stiffled by the ‘establishment’ of British Idealism. Hegel, to them, became the symbol of everything that was wrong with Britain, not only in philosophy, but also in politics, ethics and religion. It was with the first world war that they came to dominence, when the bloodshed seemed to have discredited the Hegelian path.
The founder of Analytic Philosophy was GE Moore – legally named “George Edward”, he hated his names, and was known professionally only as “GE” (his wife, unable to cope with calling her beloved by his initials, took to calling him “Bill”). It was his belief that Hegelianism was not only slipshod and illogical in details, but also profoundly wrongheaded in intent. Following the lead of the old “School of Common Sense”, Moore argued that the meanings of common language sentences are readily apparent to everybody who uses them, and that, furthermore their truth cannot be denied while using language. His most famous example is his proof that he had two hands that existed outside his mind: he waved his hands in front of his face, and said “look, I’ve got two hands”. Since we all know that “I’ve got two hands” refers to hands that exist outside our minds, and since we can all see that he has two hands, it must be the case that he has two hands that exist outside his mind.
The audacity of this argument is not unusual for Moore. In metaphysics, he claimed that not only objects but also propositions themselves were real, physically existing objects. This directly opposes the Idealist argument that only one thing exists with the opposite claim, that millions of things, perhaps even an infinite number, exist. Indeed, he at times went further, saying that only propositions existed, and not objects at all. This propositions need not be true in virtue of anything else, but simply ARE true or false – Russell expressed it by saying that propositions are true in the same way that roses are red.
Most strikingly of all, he argued that looking at pretty artworks was the basis for all morality. This, he said, could not be argued for – “good” was unanalysable, and could not be identified with either physical or metaphysical properties. Instead, we had to rely on “moral intuition” (a part of common sense) – and it was clear that intuition said that enjoying art was the highest good. From this basis, Moore manages to recreate a surprising amount of common sense morality: with delightfully bourgeois ingenuity, he argues that murder is wrong because it is inconveniently distracting – murder occuring makes people more fearful about murder, and if I am constantly going around being concerned about being blown up by anarchists I can’t properly enjoy my fine wine and my collection of rare Picassos. In order to enable art-owners to appreciate the beautiful objects they possess, the rest of us ought to jolly well keep quiet and not make too much disturbance.
Moore’s dramatic style of positive argument has not been greatly respected. Nagel, for instance, speaks of three reactions to the chasm of skepticism: the heroic, the tragic, and that of GE Moore: the heroic philosopher attempts to leap across the gap, the tragic philosopher sits and weeps because he knows he can never cross, and GE Moore walks up to the gorge, turns around, and loudly claims that he is now on the other side. Moore’s negative work, however, although now considered almost entirely wrong, has been greatly influential as a model for later philosophers.
Most importantly, Moore has been taken (perhaps contentiously) to mark the beginning of what is known as “the linguistic turn”: roughly, a turn from thinking in terms of concepts to thinking in terms of words and meanings. Making use of the idea that two different sentences can mean the same thing, Moore sought to ‘analyse’ philosophical problems by rephrasing in other, simpler and more immediately understandable, terms. This linguistic analysis was widely thought to be able to cut through all previous philosophical confusions.
Moore’s influence, though, was soon eclipsed by his colleague, Bertrand Russell, to the extent that he is now usually left off the obligatory trinity, replaced by Russell’s co-opted predecessor, Gottlob Frege, and by his appointed successor, Ludwig Wittgenstein. Russell’s chief divergence from Moore is over the nature of analysis: where Moore rephrases things in ordinary language, Russell believes that this lacks clarity; instead, he analyses sentences into an “ideal language”. This language is identical both with formal logic (for which he drew on Frege and Bradley) and with mathematics in its purest sense, and is most systematically put forward in his 1910 “Principia Mathematica”. Here, Russell applies to mathematics the same principles of linguistic analysis: he seeks to “formalise” mathematics by demonstrating its truth in a more logical and undoubtable fashion on the basis of clearly stated axioms and rules of inference (in particular, on type theory – certain axioms regarding “sets”, engineered to avoid particular paradoxes inherent in Fregean set theory).
Far smaller, and yet more significant for philosophy, was his 1905 paper, “On Denoting”, in which he rejects Moore’s “object theory of meaning” – the theory that the meaning of a sentence is the object or state of affairs that it references. This theory had produced problems when philosophers considered non-existing objects, like invisible pink unicorns – in the claim “unicorns do not exist”, are there some ghostly unicorns that exist that we are talking about, that fail to exist even while they exist? Russell’s solution was to “analyse” such claims as general statements about the world: “X does not exist” becomes “it is not the case that any thing, Y, exists, such that Y is identical to X”. Similarly, he defines “the” in a novel way: “the X is Y” means “there is a thing, Z, such that Z is X, and for all A that are X, A is Z, and for all Z, Z is Y” – or, more simply, “the tree is tall” means “a tree exists, only one tree exists, and all trees are tall”. Thus, statements are converted from claims about specific entities to claims about the world. One famous consequence of this is that all statements about non-existent entities are false: “the unicorn is tall” is false, because the claim that “a unicorn exists” is false. [Statements claiming that non-existent entities do not exist are an exception, and can be true, because “existence is not a predicate”]
It may appear that this is nonsense: when we claim that “the tree is tall”, we aren’t saying that there is only one tree, or that all trees are tall. Russell, however, believes that we ARE saying that, we just don’t realise it. In this way, Russell abandons Moore’s commitment to common sense meanings: the real meanings of what we say may be quite different from what we think we are saying. This, after all, is where philosophical confusions come from: we naively believe that when when an MP says “the honourable member is correct”, it is possible that he may be correct – but of course, as there is more than one honourable member, that claim must always be false.
Both these works exemplify the method of “ideal language analysis”; both also work toward a new metaphysical theory, known as logical atomism, which did not however come fully into coherence until after the publication of the Principia. Logical atomism is often considered the distinctive creation of Russell – but in fact Russell himself acknowledged that he was only attempting to explain certain ideas taught to him by his “pupil”, Ludwig Wittgenstein.
Wittgenstein was born to one of the wealthiest families in Europe, but this probably did not translate into a peaceful childhood – three of his four brothers killed themselves. His childhood hero was the physicist, Ludwig Boltzman, but he killed himself before the two could meet. Nonetheless, Wittgenstein became a physicist and engineer, before discovering mathematics – and in particular the formal mathematics of Frege and Russell. Through communication with Frege, he was advised to seek to study under Russell, and accordingly the young man turned up unanounced one evening at the door of Russell’s rooms in Cambridge in 1911. Within months, Russell abdicated from philosophy, on the grounds that having seen Wittgenstein’s genius, " I could not hope ever again to do fundamental work in philosophy." Wittgenstein had had no formal training in philosophy – indeed, he never had any academic degree, except for the PhD granted to him by Russell and Moore long after he had become famous – and was notably ignorant about the philosophers of the past, with the notable exception of Schopenhauer, whose project he seems to have seen himself as continuing. He had little interest, in fact, in academic philosophy, and spent only a few years as a professor – much of his time was spent in a hut in Norway, or as a schoolteacher, or an architect, or a hospital porter, or a gardener. He briefly considered becoming a monk (they wouldn’t take him), or a manual labourer in the Soviet Union (nor would they), and he gave away all his money – to the rich, so as not to further corrupt the poor. As a professor at Cambridge, he made a point of banning from his classes any student who appeared too interested in them; when members of the Vienna Circle discussed his work in front of him, he turned his back and recited Bengali poetry. Nonetheless, he is widely considered the greatest philosopher of the century.
Logical atomism, as epitomised in Wittgenstein’s “Tractatus Logico-Philosophicus” (named in honour of Spinoza’s Tractatus Theologico-Politicus; like Spinoza’s masterpiece, the Ethics, Wittgenstein’s Tractatus is presented in numbered bullet points in the style of a mathematical proof) states that all meaningful propositions can be reduced to a number of “atomic” propositions, joined together by truth-functional connectives (the Principia Mathematica admits several such, but Wittgenstein reduces them all down to the single Scheffer Stroke). Atomic propositions themselves represent atomic facts – and atomic facts are simple (irreducible, non-complex) states of affairs that link together different “objects”. These objects have substance – that is to say, they remain present and the same in all possible worlds and cannot not exist. The atomic objects are unchanging – what changes are the facts that connect them. If, for instance, my house is green, there is a certain fact relating the house and greenness, and a fact relating the house and me. These facts are reflected by atomic propositions in ideal formal language: say, “Fab” and “Gac” [although it should be noted that my house, and probably greenness, is not actually a simple object – I pick the example purely to demonstrate the concept more clearly]. These atomic propositions are then combined into a “molecular” proposition, “Fab^Gac”. The molecular proposition is then represented by words in an actual language. For a sentence to be meaningful, it must represent a proposition, and that proposition must either be elemental or must connect elemental propositions by truth-functional connectives. Those elemental propositions must then represent an atomic fact. If any part of this chain is broken, then my words are meaningless.
Alongside this is an important distinction between “showing” and “saying”: what a proposition “says” is only a truth-condition, but what the proposition “shows” is the fact itself: a proposition is a sort of “picture” of a fact, and similarly a thought is a sort of picture of a proposition – there is an isomorphy between the internal structures of the three. A proposition, Wittgenstein says, “shows” what things would be like if it were the case, and then “says” that it IS the case. In this image, we may see a return to Hume’s Fork - all meaningful claims are about matters of fact or relations of ideas – but shows a way in which the same proposition can address both simultaneously, by showing one thing while saying another. Many things cannot be said, but only shown – such as the principles of ethics and religion. These things are strictly meaningless and senseless, but not necessarily nonsense – it is only true nonsense if it has the form of something that claims to be meaningful. Unexpectedly, Wittgenstein includes all logical and mathematical propositions in the category of “meaningless” things – they only “show” us the structure of concepts, and do not “say” anything about the world (a linguistic reincarnation of Hume). The propositions of the Tractatus are therefore themselves meaningless. Accordingly, we should abandon them once we have understood them:
“My propositions are elucidatory in this way: he who understands me finally recognizes them as senseless, when he has climbed out through them, on them, over them. (He must so to speak throw away the ladder, after he has climbed up on it.) He must transcend these propositions, and then he will see the world aright.”
As a concrete example of “throwing away the ladder”, we may consider the problem of Tractarian objects. The Tractatus says that they exist – indeed, that they must exist – yet it also implies that existence is something that is only meaningful in complex propositions. Strictly speaking, therefore, a Tractarian object is “that for which there is neither existence nor non-existence” – when we say that they exist, we are using a different meaning of “exists” from that used ordinarily to say, for instance, “the car exists”, because “the car exists” is ultimately cashed out in terms of the relations of particular simple objects, and a claim about one simple object by definition cannot be reduced to any relation between simple objects (else it would not be simple). Therefore, the claim that simples exist, although fundamental to the Tractatus, is a demonstration of trying to “say” what can only properly be “shown” – it is meaningless, but it informs and shapes our thoughts. This is an Analytic return to Schopenhauer’s views on the world-in-itself, knowable only through flawed but indispensible perspectives, beyond even being and non-being.
Another example of the meaningless is inductive reasoning, which Wittgenstein, following Hume, insists is only “psychologically” valid, not actually so – because induction is “reasoning” on the irrational basis of what the picture of our experience shows us, not genuine, linguistic, reasoning on the basis of what our experience says. What is shown is beyond logic, because logic is a set of rules governing what should be said.
In this respect, it is important to note that Wittgenstein had a problematic relationship to religion, both Judaism and Christianity – although he was not a believer, he did sometimes consider himself either a Jew or a Christian in a religious sense, and he had great respect for religion. The Tractatus was inspired not only by Russell and Frege, but also by his reading of Tolstoy and Kierkegaard; in the trenches, he was known to his fellow soldiers for handing out religious tracts, and called “the man with the gospels”. Consequently, we should not imagine that when he rules religion, or ethics, or mathematics, “meaningless”, he necessarily considers them unimportant. Indeed, the Tractatus ends with the gnomic intonation: “7. Whereof we cannot speak, thereof we must be silent”, and this was intended to be the foundation of his new life: the Tractatus, indeed the whole of philosophy, was only a preface, to be dealt with quickly before turning to the things that really mattered – the realities of human experience, or what he called “the mystical”.
The Tractatus was sent to Russell from an Italian POW camp, five years after they last communicated with each other, at a time when Russell did not even know that Wittgenstein was still alive. With the exception of one small article, it was the only work that Wittgenstein ever published.
Wittgenstein’s immediate influence on philosophy was immense, and he became one of the few 20th century philosophers to be genuinely famous in his own day. Although the details of his logical atomism are not now widely accepted – and were indeed entirely abandoned by Wittgenstein himelf later on – his programme helped to set the course and style of analytical philosophy for half a century. Wittgenstein’s theories allowed sharp dividing lines to be drawn between the philosophy of saying and the unaccountable, irresoluble, speculative pseudo-philosophy of showing – the former was to be the province of the “analyst”, calm and scholarly and usually collaborative, while the latter was the realm of the metaphysician, obscurantist and attention-seeking, typified in history by Hege and in the modern age by Heidegger.
Wittgenstein, however, faded into the background after the publication of the Tractatus, due to his refusal to publish and his reluctance to teach. Instead, he was succeeded by a school known as Logical Positivism; this was typified by the Vienna Circle, a philosophical discussion group that centrally concerned itself with Mach, Einstein, Russell and Wittgenstein. The group varied in constitution over the years, but key members included Otto Neurath, Moritz Schlick, Rudolf Carnap and Friedrich Waismann; they were popularised in Britain by AJ Ayer, and had the support of many of the prominent members of the original Analytic movement, including the mathematician and logician, Frank Ramsay, and the economist, JM Keynes; Kurt Gödel was an avowed member of the circle, although he disagreed with the other members on several issues, while their compatriot, Karl Popper, was published by them and in turn helped publicise their work; the American, WVO Quine, attended meetings on his travels around Europe.
Logical Positivism follows the Comtean heritage through the scientific lens of Mach: it believed in the importance of positive science, conducted through experimentation. From its Empiricist heritage, it inherited the dogma that all knowledge comes from experience, and all experience comes from the senses, which meant that “experimentation” had to have results that could be distinguished by the senses. Like the Pragmatists, they believed that the meaning of a concept (or word) was its ‘practical consequences’, but because of their positivist/empiricist heritage, they interpreted this to mean testable sensory consequences, and like Wittgenstein they moved the emphasis from single concepts to propositions. If the consequences are not testable, they are not practical, which means that testability becomes central to meaning. For example, “diamond is harder than glass” can be tested by scratching one against the other and seeing which is damaged. There is no skeptical gulf here – the test does not merely provide evidence for the proposition, but actually prooves the proposition (we can continue to doubt that the test has been performed accurately, but if we grant that it has, its results are definitive). It is important to note that a test is not an object, but rather a procedure – this gives us “the verification theory of meaning”, according to which the meaning of a claim is the procedure by which we test its truth, which includes the “truth conditions” – the sensory data we will receive if the theory is true. Essentially, all propositions are now to be seen as scientific hypotheses.
One consequence of this is a return to Millian phenomenalism. Because nothing can be said about an object that is not a proposition, and each proposition is testable, and each test identifies verifying sensory criteria, this means that the object itself is identified only through sensory criteria. We cannot talk of having all the sensory evidence to show that something is glass and yet doubting that it is glass – if the sensory evidence is there, it is glass. This evidence cannot practically be assembled, because it probably involves a whole range of hypothetical sensations we receive from running a large, perhaps infinite, number of tests, but nonetheless the object has in theory no criteria other than the evidence. The object can thus be considered to BE those ‘permanent possibilities of sensation’; we cannot object “but there’s more to it than anything we can sense”, because we cannot, through our senses, test whether there is anything we cannot sense. The ‘thing-in-itself’ is, as in Kant, excluded from our thought, but because we have taken the linguistic turn it is also excluded from our language – which means that we cannot be talking about the thing-in-itself at all, and if we insist that we are we are deluding ourselves. All there is is evidence, and all evidence is evidence for the hypothesis that we will receive such-and-such other evidence if we do such-and-such; all there is is evidence, and there is nothing ultimately ‘behind’ the evidence. It is illogical to even suggest that there may be – the words literally have no meaning, and we cannot think what we may think we think.
This phenomenalism/empiricism is then imported into the Tractatus: the Circle interpreted the Tractatus as talking about elementary sensations when it spoke of objects, with those sensations combined into ‘facts’. With the Tractarian scythe, the Circle ruled that all things that were not “saying” something about real or possible sensations were meaningless – and they were not as generous toward the meaningless as Wittgenstein himself was. Accordingly, all ethics and metaphysics were to be eliminated – although they were puzzled by the way that Wittgenstein extended meaninglessnes to the Tractatus itself. They accepted that the Tractatus, like the rest of logic and mathematics, was meaningless in the sense of not describing the world, but only our concepts – but nonetheless considered it important and true, not in a propositional sense, but as a series of guidelines for action. As an example: according to the verification principle, universal statements (“all (actual and possible) sheep are mammals”) are meaningless, because they cannot ever be verified; however, Wittgenstein tells us, and the Circle (initially) accepts, that universals can be taken as guidelines for the formation of particulars. So, “all sheep are mammals” tells us “if X is a sheep, it is legitimate to form the claim ‘X is a mammal’”; however, it does not SAY that we may do this (because it cannot), but the form of the universal SHOWS us the forms of the particulars that we may construct.
The result of this is scientism and the doctrine of unified science: all knowledge comes from science, and all sciences are only branches of one science. Results in any scientific discipline must be translatable into results in any other: if, for instance, psychology is to be meaningful, it must be translatable into verifiable theories in neurology, and ultimately in physics. Meanwhile, the task of philosophy is to deal with the language through which science is conducted and communicated: although this linguistic effort is strictly meaningless, it is important to render it as unentangled as possible so that the meaning of the science itself is more accessible. Philosophy therefore has two objectives: to winnow out all claims that are not scientific, and to analyse those that are into verification criteria that scientists can then devise tests for. Logical positivism is therefore all about “operationalisation” – it itself is an operationalisation of Russell and Wittgenstein (showing what their work means for practical science) and it believes that all philosophy should assist in operationalising the otherwise vague claims of language.
Because ethics, religion and so forth are rendered meaningless, in their conventional senses, and yet are clearly important to people, logical positivism had to provide some other account of their import. This is the doctrine known as “non-cognitivism” or “emotivism” – the surface form of statements in the domain of X (eg ethics or religion or aesthetics) may mirror the surface form of propositions of science (eg “murder is wrong” looks as though it ascribes a quality, “wrongness” to an entity, “murder”), but when such statements are analysed we see that the underlying form is really quite different. “Murder is wrong”, for instance, is actually an expression of an emotion: “Murder? Ugghh!” (for ‘emotivists’) or “Don’t kill people!” (prescriptivists) or “I acceed to the established social rule against killing people” (norm-expressivists).
Some logical positivist – most notably Carnap – attempted to validate the method of induction by describing “probability” not as a physical property but as a logical/epistemological one: under his ‘confirmation theory’ of probability, to say that there is a 50% chance of something happening means that if it does happen, a theory that predicted it happening is only 50% confirmed. Theories from induction, therefore, may never be entirely confirmed, but can be confirmed to increasing degrees by the mounting up of evidence. From this, Carnap attempted to build a formal “inductive logic” that can incorporate and quantify degrees of confirmation – unfortunately, despite the work of many years, he discovered that in every formalisation he could create, all universal laws were always 0% confirmed. Late in his life, he began again, attempting to address probability (and hence induction) on the basis of thermodynamic entropy, but he died before getting very far.
Logical positivism was an enthusiastic attempt to return to sure footing. On the basis of “given” sensory evidence and linguistically inescapable rules, it attempted to construct logically and formally an account of what could be known – all those things that could not be arrived at from this foundation could be ‘cast into the flames’. In this way, logical positivism was a restatement of Humean skepticism, and Hume’s Fork: all knowledge is of matters of fact (now refigured as the sensory verification of positivist hypotheses) or of relations of ideas (now reconfigured as the structure of language, and of the ‘ideal language’, formal logic, that underlay it); relations of ideas tell us nothing about the world. Fittingly, one of the most deadly blows against the school was from an axe left lying by Hume: induction. A great deal of what we take for granted is not available without induction, and induction cannot be justified from experience, nor formalised controllably. In particular, the principle of verification itself was imperilled, because of its inductive subtext: verification criteria only apply in “normal conditions”, a background state of affairs created by inductive reasoning; we have no way to formally know which conditions we can take as normal and enduring (and hence not consider relevant to our hypotheses), and which are fleeting (and hence must be considered); moreover, even if we did know this, we have no way of specifying them, as the possible conditions are not only complex but formally infinite.
Moreover, the logical positivist dependence upon science became vexing when science failed to meet the standards the school demanded. In mathematics and logic, it became increasingly difficult to construct sufficiently complete and consistent logical systems to model the sort of natural language and inference used by scientists (Carnap innovatively attempted to escape by allowing multiple, even inconsistent, logical systems); in experimental physics, scientists were increasingly turning to unverifiable and indirect theories and meta-theories, often lacking complete real interpretations, and it proved impossible to distinguish endeavours such as quantum mechanics from the old-fashioned metaphysics or theology that logical positivism had attempted to exorcise. Finally, in the work of Quine and Sellars, core precepts of empiricism itself were called into question. Combined with the work of the Later Wittgenstein, this caused logical positivism to slowly die away after WWII.
But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!