Idealism & Russell’s Metaphysics

Part of my slow working through Mumford’s “Russell on Metaphysics“, I have flipped back to recap the first 5 chapters. Having been left with even deeper nagging doubts that I was right to leave Russell behind after all, I need to check my thinking by capturing the key thoughts. A sanity check.

To be clear, I can summarise already that, having now read a lot more of Mumford’s collection and commentary than just that noted above, I can see how Russell did indeed engage in metaphysical thinking along the way to his analytic philosophy founded on realism. That is NOT at issue.

What I’m suggesting is that the philosophical position he arrived at is flawed – detached – metaphysically. In fact my main thesis is that this realist analytic philosophy – what I’ve variously referred to as a logically, materially and empirically objective positivism or scientism in my own mental travels – is more problem than solution to the big questions of our time. And, as I’ve said I’m less concerned with definitions of isms and what they are called (after say Feynman and Dennett), than I am with understanding specific sets of assumptions and assertions.

Hence my unpicking particular problematic assertions below.

Background note:
This exercise started here – A New Leaf
And continued here: Pan-Psychism
Here: Doubts on Russell & Idealism
And here: Dual Aspect Monism

Although I’m focussing on Part 1 (the first 5 chapters), most (but not all) of what follows for now actually uses Mumford’s introductions – to the whole book and to the specific chapters.

Firstly I need to say that Mumford’s little introduction to Metaphysics is itself tremendously valuable. To recognise the distinction between the metaphysical aspects of philosophy and the empirical aspects of philosophy and the sciences. In many ways it is this central issue that most concerns me here. Recognising their distinct natures is fundamental, but having used one to establish the other doesn’t mean we can thereafter dispense with the former – throwing the metaphysical baby out with the scientistic bathwater. The real world still cares about metaphysics and relationships with the physical, recognising the proper value in the relationship of the non-empirical alongside the empirical.

“Russell’s major and lasting contribution to metaphysics has been hugely influential and his insights have led to the establishment of analytic philosophy as a dominant stream in philosophy.”

Mumford’s editor.

“Contribution to metaphysics”? Sure, but it just seems to have been a means to his end: the “[dominance] of analytic philosophy” – precisely the problem that concerns me.

His later

“Starting with papers from his pre-analytic period, the volume collects Russell’s main realist accounts, his discussions of the problems of universals, and his writing on causation and the laws of nature. The final part covers Russell’s thoughts on diverse questions of metaphysics that occupied him later in life, including his classic paper on vagueness.”

Mumford’s editor

OK, fair enough. I’m focussing on his pre-analytic metaphysical journey here, to realism from idealism. Causation, free-will and the laws of nature (and time) are also fundamental to my own interests (See Unger & Smolin and Rovelli and many others cited here.) Ditto language and vagueness – I’m a fan of Wittgenstein, Dennett and more after all – all covered elsewhere and no doubt further in future, but not mentioned explicitly again in this post.

“Russell achieved public fame – often enough, notoriety – because of his engagement in social and political debates, becoming known to a wide audience as a philosopher in the popular sense of the term.”

Grayling’s introduction to the Russell series.

Again I mentioned in the preamble posts, I really do appreciate Russell as a fully engaged public intellectual – campaigning in skeptical, secular, rational, humanist areas of life, like myself I like to think. My issue with Russell is his flawed metaphysics, which obviously informs the detail of some of his ideas in real life.

[Meta – internal referencing of Mumford’s book can be a little confusing. As he acknowledges, many of the Russell papers and extracts are taken from “The Collected Papers”, so we have the book with it’s own chapter headings and numbers, the papers as numbered in the source collection, and the references Russell used withing the selected writings. It’s complete and thorough, just keeps you on your toes in following the journey.]

“[Russell was] one of the greatest analytic philosophers of the twentieth century”


Not in doubt. It’s the analytic philosophy that is the problem (for me).

“there is no better introduction to metaphysics”


Yes, as I say above, that has real value here in its own right.

“To some who know a little of Russell’s philosophy, it might seem strange to speak of him being engaged in metaphysics. He is often depicted as standing squarely in the empiricist tradition that had, on the whole, rejected metaphysics and was concerned primarily with the theory of knowledge or epistemology. If this book has but one aim, it is to relieve its readers of that misconception.

Russell was a metaphysician.”


Yes, I quoted this already in an earlier post. I am Mumford’s target audience and I consider myself having been relieved of that particular misconception. My problem now is not whether he engaged in metaphysics, but that his eventual metaphysics was flawed.

“British Hegelian idealism [had evolved by inheritance from Kant > Hegel > Green > Bradley]. Russell [and G. E. Moore] rejected idealism for metaphysical reasons.”


Fair enough. What is therefore crucial to my thesis is what was Russell’s understanding of idealism at the time and where was his basis for rejection, and his adoption of realism, flawed in the long run.

“Thought is a pre-condition of knowledge in the way Kant set out.”


Oh boy. This is a definitional, almost tautological statement, but whilst I don’t deny Kant said this – as we’ve discovered before – words and concepts like knowledge don’t travel well in translation [see summary from French for example, below in final paras]. To be known requires some mental activity or state. To be knowable requires some information (or disembodied pattern of significance) to exist. Knowledge requires both. This will recur, and it’s why I shift my metaphysical and epistemological focus to information. [And obviously this is already a lead-in to idealism as something maybe, kinda, sorta pan-psychic, if we were to somehow privilege the mental over the (otherwise real) physical.]

“‘Appearance and Reality’ (Bradley 1893) [was described positively] by Russell [and Moore] as ‘epoch-making’ in 1895. By 1897 both had come to reject [the whole idealist tradition].

Bradley had argued that neither immediate experience nor relations are real, and from the latter we can infer that no ordinary phenomena are real and that there is no absolute truth or falsehood. Reality is a single togetherness rather than being many distinct, related things. The things that appear to us as distinct individuals are actually aspects of the comprehensive, concrete individual, which Bradley calls the Absolute.

In rejecting Bradley and idealism, Russell and Moore came to be realists. They accepted as real all the everyday, common sense, things that Bradley had told us were mere illusions. Whereas Bradley wanted to push appearance and reality apart, Russell and Moore sought to bring them together.”


It’s easy to reject this version of idealism, described as Bradley’s here, that’s for sure. The whole Appearance vs Reality saga is long-established and my readings made good use of Owen Barfield’s “Saving The Appearances“. But the point is what about it is being rejected and what makes realism better in those particular respects?

[Could already say a lot more here: The “real” of realism indeed sounds a lot more like common sense, but that would be to ignore the definitions of the real and ideal (and their relations) actually being used here. But the idea that there might be something more fundamental – more absolute – than either physical reality or the appearances experienced is not necessarily what is wrong. What is wrong is the idea that it must be one or the other – the suggestion that this is all there is. The pushing apart vs pulling together difference is key – without privileging either – we are dealing with at least both and a relation. My position is there is a third kind, more fundamental than either and the two kinds being debated here are both evolved from this – a dual-aspect-monism. I’ve been calling it pan-proto-pscychism as a reaction to the prevailing (dominant) paradigm to restore the mental to the same level as the (physically) real, but in fact the proto-component (information, “res informatica”) underlies both real/physical and ideal/mental, so I can see why people reject privileging the psychism as the name of this position in a non-reactionary sense. Will read on before coming back to unpick this.]

What is interesting, in Mumford’s overall introduction from this point onwards, is how quickly Russell moves from rejecting Bradley over 2 years and need to refute his idealism, leading him in only 3 further years to developing his new Principles of Mathematics using the new logics of Frege [and Cantor].

It’s as if having refuted idealism, realism must be right, so without further ado let’s construct our new ontology of existence.

“The new logic had a metaphysical basis, however. It assumed all sorts of things that Bradley had rejected. It assumed certain objects such as real and mind-independent propositions. It assumed objective truth and falsehood, regardless of belief. It assumed the existence of relations with an independence that was external to their relata. It also assumed a plurality of objects. Russell discussed some of these questions of ontology in the papers collected in Part II. An ontology is simply an inventory of what there is. For a metaphysician, this will be a list of the categories of things that exist, such as propositions, properties and relations.”


Telling is that Mumford’s example ontological categories includes neither real/physical nor ideal/mental stuff? Particularly scary is the idea of “mind-independent propositions” – proposed by who? I ask. But this introductory stuff if getting beyond the Part I on the idealism > realism turn.

[And … adding Frank Ramsey / Wittgenstein linguistic turn to this. Not forgetting Grayling’s reminder on Ramsey himself.]

“Metaphysics then endeavours to bridge the chasm between physics and psychology. Seeing that the objective reference of an idea is known as intuitively and immediately as its subjective nature (if not more so), it frankly accepts both; it allows a world other than the individual mind, concerning which we have knowledge and desire; its criticism is not of these them selves, but of a world concerning which they are possible: given Self and the world in relation, the problem is to make each term and their relation intelligible; and Self and the World are given, because the only alternative is blank and absolute scepticism.”


Seems fair enough. Particularly like the suggestion of the alternative as a “blank and absolute scepticism” – a recurring “scientistic” problem of everything being open to doubt if no metaphysical bootstrapping is accepted. A world of only empirically objective evidence is …. dead.

Finding it hard not to jump straight to my own position again, which I’m guessing turns out to be Russell’s once I get it. What I’m finding is that the either/or debate between Idealism and Realism is “fake” – neither can exclusively be the case in any narrowly defined sense. What matters is how they are related, and whether either has priority in any useful sense. In fact the Foucault triad seems the most sensible conclusion [See later].

That is: We can postulate a real-world “out there” – objective and mind-independent “facts”. But there are only two things we can actually deal with and that isn’t one of them. They are (1) Our empirical experience of it, and (2) the conceptual model of it (and our psychology) which we construct and revise as we evolve.

That fact we can never get a handle directly on the out-there real-world, is no reason to say it doesn’t exist and that only the ideal world exists. Idealism is clearly wrong, defined that way, but literal realism can never be be any use either, without accepting that we can only ever deal (in practice) with our experience of it and our model of it. A kind of pragmatic realism.

Everything else – the linguistic turn – says we might choose a limited (say, predicate) logic for our formal model of supposedly real-world facts, but in fact (a) we can choose (and evolve) the set of rules, logics (and exceptions) for our model (ontology), and (b) the relationship between our experience and our model cannot in fact be limited by arbitrary linguistic rules imposed by the model anyway (Russell’s paradox and Godel’s incompleteness, etc). Information and communication are rhetorically creative mental gymnastics – word games – between our experience and our model. And, it is almost impossible in the normal course of life to keep these two entirely distinct. The absolute – radical empirical – experience is ephemeral and almost immediately interpreted through our psychological model – the imperfect, incomplete and necessarily biased version of the formal model held in our minds, both conscious and subconscious.

Ha, right on cue. Russell attempts to impose formal rules on Wittgenstein’s language of reality. (Everything I see confirms my own take on Wittgenstein – the frustration that Russell never got the point of his Tractatus. Also neatly sums up my problem with modern scrabble as an “evolved game”):

I’m not sure labelling this ability to mentally hold both a model and an experience of the real world as compatibilist or reconcilliationist really helps. It’s not a problem to be reconciled, but a fact of life that the three-way relationships need to be acknowledged and understood pragmatically.

[Going to cut and run from this post …. with some final draft thoughts for now. Not a proper review, needs some significant editing.]

“Trialism” – a triad view of reality, or dual-aspect monism (after Foucault). A proper resolution of idealism (pan-psychism) and realism (physicalism), not simply a compatibilism?

Any one of the nodes is a view of the whole, but depends on the other two perspectives. The arrows are two-way interactions.  The axes are spectra, convenient dividing lines, but largely gradual “more or less” scale – which means the physical includes the pan-proto-psychic (everything is a combination of both). No reason to pick sides between physical reality and pan-psychism.

[Convenient dividing lines on gradual scales? = “good-fences”.
See also defining “identity”.]

[Post Note: Ray Tallis is not a philosopher I often find reason to agree with – unless I’m conflating experiences I have found him at the scientistic end of the spectrum. However this review of his “Logos” had one phrase that jumped out at me – “knowledge is a relational property” – which seemed to be a fundamental feature of my diagram above. (h/t David Morey for the link).]

3 thoughts on “Idealism & Russell’s Metaphysics”

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.