Home

Papers

Blog

Pictures

About Me

Contact

GitHub

KISS Linux


Critical Review - Bertrand Russell's 'Philosophy of Logical Atomism'


Bertrand Russell's 'Philosophy of Logical Atomism' is a collection of lectures
Russell gave in the early 20th century. Logical atomism offers a
strong divergence from prevailing British Philosophy at the time, namely the
long-held position of idealism. In this work, Russell aims to explore the nature
of reality as it truly is, and not as some sort of relation between minds and
projections onto an external world. The first seven sections provide arguments
and explanations of linguistic meaning and how it relates to logical atoms. The
eighth section explores more metaphysical questions that, while Russell rejects
them, sees them as being relatively important to the kinds of questions we ask
and answers we explore. Finally, the collection includes a paper Russell wrote
in 1924 on Logical Atomism proper, dealing both with his own philosophical
upbringing and his views on knowledge, and how these relate to his position in
favor of the existence of logical atoms and their relationship to linguistic
meaning. Russell holds a strong realist position and seeks to carve reality up
at its proper joints, arguing that perceptions and meaning are deeply
intertwined and that the ways in which we speak about the world line up with how
it actually is. However, Russell does reject calling his philosophy a 'realism',
saying in the final essay instead that his position should be and others should
be characterized by their logic rather than their metaphysic, and that he could
happily change his view on the issue without changing what he believes about his
doctrine. His position can be construed as one of a rationalist or an
empiricist; indeed, Russell seems to argue for his logical atomism on both sides
of this historic division. It isn't altogether clear which we should find more
convincing, but perhaps we should take them both for what they are: strong
support for his position for any sort of philosopher. His shyness about realism
may in part be to distance himself from criticisms of his view, which he readily
admits exist, which I shall discuss later. Altogether, while Russell espouses an
affirmative position on the existence and analyzability of true logical atoms,
something is problematic in that the question of whether such an analysis can
end, whether it be in principle or in practice, is still left unsaid in a
rigorous way by Russell.

Russell's arguments for logical atomism can be considered in terms of the
primary question of how it is we understand the world around us. Russell
believes that our navigation of the world and our interactions within it are
guided by our use of language. The primary concern before us, then, is whether
or not our language properly expresses facts. That is to say, does our language
describe the world correctly, and is there a correspondence between language and
the world? The question of how is it that our language corresponds to the real
world in any capacity has been answered several times over. Locke offered the
position that meanings are relations on ideas that we have in our mind to the
objects we are speaking of. Hume suggested that our language is a projection of
our categories onto the world. The contrast with the idealist, Russell seeks to
answer the question of whether or not what we mean could line with the
fundamentals of reality, even through the mediation of our thoughts. In short,
will what we mean be clouded by how we perceive the world, or can we actually
capture reality? The paramount argument for Russell's idea is that, supposing
our language did not properly line up with the joints of the world, the things
that we say and think would all be horribly mistaken and radically incorrect. We
would, presumably, prefer that what we say and think does in some sort of
correct way, line up with the ways in which the world actually is. Thus, our
meanings should be taken to reflect this partitioning. The entire foundation of
logical atomism appears to be an attempt to preserve what we say and think about
the world and argue that it is fundamentally true. While not much in the way of
arguing beyond this is offered up, the pragmatic reasons for accepting it seem
quite enticing. If the idealist were correct, after all, we would find ourselves
hard pressed to agree with each other on what our perceptions were, and actual
truth would be quite beyond our grasp. To Russell, it seems far simpler to argue
that what we say about the world should characterize it correctly. For if we are
wrong and our meaning doesn't carve reality closely at its joints, perhaps we
are only misunderstanding the world, and a further analysis would be required.

This correspondence between language and the world leads to the question of what
exactly is it in our language that is corresponding to the world? Is this
correspondence between complexes, things that cannot be further broken down
without losing important aspects of the whole, or can a further analysis be
performed, breaking up these complexes into finer and finer objects, reducing to
the simplest of things? For Russell, this correspondence bottoms out at the
level of simples, arrived at through further and further refinements of our
analysis of the complexes, until we arrive at the simplest things about which
our language and thoughts correspond to reality. These simples are what Russell
terms the 'logical atoms', the most fundamental constituents of our
sense-experience, the sense-data which our meaning is about. Russell argues that
the world naturally decomposes into these kinds of atoms. This can be seen when
we attempt to analyze our words in our language. We start with a word and look
up its meaning, analyzing it into the things that make it up, it's constitutive
parts in our language. From this analysis we continue, going further and further
down in our language, until we find ourselves bottoming out at the simplest
things, those which compose all our words and our thoughts, the things we cannot
seem to find a simpler part of. We cannot further analyze these terms, and thus
they are atoms, indivisible by any further sort of analysis.  Russell seems
quite convinced that this is the case. Indeed, his argument can be traced back
to a similar one from Leibniz, lending credence to the position that Russell's
argument is one of a rationalist bent. The argument follows the lines that
complexes are such insofar as they are composed of simples - if a complex is
taken to be what it is (namely, a composite), then it seems clear that such a
thing can be broken down further, into simpler pieces. However, this leaves
quite the serious problem for Russell. For if it is the case that we can break
complexes down further and further in analysis to their constitutive simples,
what right do we have in supposing that these simples are the atomic ones? How
do we determine that our analysis has ended? Russell himself finds it obvious to
be the case that this analysis is possible, but does point out that the number
of constituents may be infinite. Russell runs into what may be considered a more
fundamental problem with logical atomism: how does our knowledge of our language
impinge on our analysis of complexes into simples?  Perhaps simples are merely
the barest words relating to sense-experience, but this only speaks to general
human limitations on our experiences of the world
- would it be correct for a logical atomist to say that for beings which share a
  similar experience to ours would carve the world out in similar ways to us,
and those beings which have strikingly dissimilar experiences of the world an
altogether different language corresponding to it? This idea seems to entirely
miss the purpose of logical atomism; it is not specific to sense-experience, but
instead specific to the world itself. For if our meaning and thoughts were
merely about our sense-experience and sense-datum, it seems that we may well
indeed be confused about what the world is, and our language is mistakenly in
its meaning of its words. 

The rationalist account of logical atoms seems to not provide any kind of proper
resolution to this problem. As far as it seems, Russell is able only to say that
complexes compose simples, but not that these simples are in themselves not
complexes. Russell thus seems to make a shift in his arguments, instead opting
for a more empirical approach to the question. Russell argues that the analysis
we perform depends upon direct acquaintance with objects. These objects are what
our meanings correspond to. These objects are not themselves experienced to be
simples, but are such in virtue of what we conclude through our logical
analysis. That we cannot break it down any further is the reason for their being
simples. It seems clear that Russell's analysis should bottom out at the level
of the world, so that our meanings truly correspond to the world. However, it
seems to be the case that, for Russell, all that truly matters is the
conclusions we draw about the logical analysis we perform. That we can no longer
carve up the world into simpler simples should demonstrate to us that our
linguistic capabilities can only go as far as that. These simples are, at the
very least, simples for us, and correspond to our thoughts and meanings. As a
result, we are not mistaken about how we are carving up the world. However, this
is problematic, if only because it means that we cannot know it to be the case
that the world cannot be partitioned any further. This, perhaps, is the reason
why Russell prefers to call his philosophy that of logical atomism, and not a
realism. Perhaps he too remains unconvinced about the depth of this analysis,
and that it can ever truly bottom out at the 'true' joints of the world,
pointing out the 'actual' natural kinds that exist. Instead, we can only ever
get as close as our own experience of the world allows. At the very least,
however, we can use this position on logical atomism to reject idealism, for it
does seem to be the case that our thoughts and meanings are deeply connected to
the world as it is.

As it were, I find myself not terribly convinced by Russell's position on how
our language relates to the world. The problem presented to us by positive
arguments in terms of performing a logical analysis seem too strong to
disregard. For instance, tt isn't altogether clear to me why we should think
that our language is a relation of simples to the world, as much of what we say
seems to go over and above the power of simples in terms of descriptions.
However, Russell does offer a quite satisfying characterization of the ways in
which we talk and how it relates to our own experiences of the world, and
grounding our language in terms of our experiences seems far more convincing
than other, more idealist arguments about the structure of reality and our use
of language. Russell's logical atomism goes through visible changes over the
years, and his explanations become more refined and better articulated in the
final essay in the collection, focusing far more on the overall pictures being
painted about how the entire theory is intended to work out. Overall, the work
is well designed and, for the most part, properly argued for, if not missing the
mark on its overall goals. 


________________________________________________________________________________

Dilyn Corner (C) 2020-2022