Is there a rhinoceros in the room?  One of the earliest encounters 
between Bertrand Russell and the young Ludwig Wittgenstein involved a 
discussion about whether there was a rhinoceros in their room. 
Apparently, when Wittgenstein 'refused to admit that it was certain that
 there was not a rhinoceros in the room,' Russell half-jokingly looked 
underneath the desks to prove it. But to no avail. 'My German engineer, I
 think, is a fool,' concluded Russell. 'He thinks nothing empirical is 
knowable-I asked him to admit that there was not a rhinoceros in the 
room, but he wouldn't.'[1]
The crux of the dispute appears to be a thesis held by Wittgenstein 
at the time concerning 'asserted propositions.' According to Russell, 
Wittgenstein maintained that 'there is nothing in the world except 
asserted propositions' and refused 'to admit the existence of anything 
except asserted propositions.'[2] But what this thesis amounts to and 
how it is related to his remarks about nothing empirical being knowable 
and about whether there is a rhinoceros in the room is difficult to 
determine. For one thing, it is difficult to see how Wittgenstein could 
be arguing that nothing empirical is knowable given the central 
importance for his early thinking of his idea that only propositions of 
natural science can be said. For another, his reported claim that there 
is nothing in the world except asserted propositions is hard to square 
with his contention in the 'Notes on Logic' that there are only 
unasserted propositions. What we need is an interpretation that can make
 sense of Wittgenstein's reported remarks, while taking into account 
their relation to his fundamental ideas and his views in the 'Notes on 
Logic' and elsewhere. Also, it must offer some account of Russell's 
extreme reaction to Wittgenstein and his worry that Wittgenstein may 
have been a fool.
In his recent biography, Wittgenstein: A Life, Brian McGuinness 
proposes an interesting interpretation of Russell and Wittgenstein's 
conversation, one echoed by Ray Monk in his Wittgenstein: The Duty of 
Genius. In what follows, I criticize McGuinness' interpretation and in 
its place propose an alternative way of reading 'asserted proposition.' 
This alternative provides us with a way of seeing Wittgenstein's 
earliest thoughts as continuous with fundamental insights expressed not 
only in the Tractatus, but in his later philosophy as well. Indeed, if I
 am right, Wittgenstein's objection to Russell anticipates ideas 
normally associated with On Certainty.[3]
McGuinness' interpretation depends on sorting out what Wittgenstein 
meant by an 'asserted proposition' and why he thought that Russell's 
remark about the rhinoceros did not qualify as one. To this end, he 
insists that we must see Wittgenstein's objection as expressing a thesis
 that is more adequately expressed in the Tractatus. This thesis, says 
McGuinness, concerns the logical composition of the world. His view is 
that 'the claim that only asserted propositions exist is clearly 
intended as a correction of Moore's position in his 1899 article [`The 
Nature of Judgement'] according to which the world is formed of 
concepts.'[4] According to McGuinness, Wittgenstein's correction is 
based on the idea that the world consists of facts-facts being asserted 
propositions-not of things or what Moore called simple concepts. The 
correction thus seems to anticipate the idea that 'the world is the 
totality of facts, not of things,' the second remark of the 
Tractatus.[5]
McGuinness reminds us that the phrase 'asserted proposition' is 
central to the accounts of the nature of a proposition defended by 
Russell and Moore, accounts that Wittgenstein is practically certain to 
have known about. The situation, as McGuinness has it, is that 
Wittgenstein had already formed an objection to Russell and Moore, which
 he then attempted to express in his conversation with Russell. In sum, 
McGuinness assumes that Wittgenstein meant by the phrase 'asserted 
proposition' what Russell and Moore had meant by it.[6]
The notion of an 'asserted proposition' is connected with Russell 
and Moore's belief that the content of a proposition is its essential 
feature and their view that the psychological processes involved in 
judgments concerning this content have a secondary status. On their 
conception, a proposition is not a psychic phenomenon as it is for Locke
 but rather is what Lockean ideas and the like are about.[7] Moore 
called the entities that make up propositions 'concepts' and Russell 
called them 'terms.' A proposition, on this view, is what Moore took to 
be a complex or what Russell called a set of terms. It is not something 
mental, but rather a complex or collection of subsistent, Platonic, 
entities.
On Russell and Moore's conception, facts are identified with true 
propositions. Truth is not-as it is on the correspondence theory-a 
relationship between a proposition (considered as a mental or linguistic
 entity) and something else. Rather it is a property of a proposition, 
now considered as a complex or configuration of terms. Some proposition 
just happens to be true, and those propositions are facts. As Moore 
says,
     Once it is definitely recognized that the proposition is to 
denote not a belief (in the psychological sense),      nor a form of 
words, but the object of belief, it seems plain that it differs in no 
respect from the reality to      which it is supposed merely to 
correspond, i.e. the truth that 'I exist' differs in no respect from the
      corresponding reality 'my existence.' [8]
What differentiates a true proposition, or a fact, from a false 
proposition is the quality it has of 'being asserted.' Russell says,
     True and false propositions alike are in some sense entities, 
and are in some sense capable of being      logical subjects; but when a
 proposition happens to be true, it has a further quality over and above
 that      which it shares with false propositions, and it is this 
further quality which is what I mean by assertion in      a logical as 
opposed to a psychological sense.[9]
An asserted proposition, then, is Russell's term for differentiating
 a true proposition, a fact, from a false proposition; true propositions
 have the property of 'being asserted,' which false propositions lack.
McGuinness thinks that Wittgenstein was harking back to this use of 
the phrase 'asserted proposition' in his conversation with Russell. He 
thinks that by saying 'there is nothing in the world except asserted 
propositions,' Wittgenstein was intending to challenge Russell and 
Moore's basic assumption that there was something more fundamental than 
facts. On the view being attributed to Wittgenstein, false propositions 
are not 'entities,' as Russell and Moore believed; there is not a 
complex of terms (or concepts) in virtue of which something is not; the 
world is composed of facts, not of terms, concepts, or things.
For McGuinness, then, the discussion between Wittgenstein and 
Russell amounted to the question 'What complex can reasonably be 
supposed to exist in virtue of there not being a rhinoceros in the 
room?'[10]  He holds that Russell was of the view that such a complex 
existed, whereas Wittgenstein in arguing that there was nothing except 
asserted propositions, was denying this claim. As McGuinness puts it,
     [Wittgenstein was] denying existence in this sense to 
everything except asserted propositions or facts.      Thus he had 
already reached the position expressed in the first propositions of the 
Tractatus that the      world consists of facts . . . [and that] things,
 objects, or what Moore called simple concepts do not go to      make up
 the world.'[11]
In spite of McGuinness' insistence that Wittgenstein's remark was 
'clearly intended' as a correction of Moore's position, we must surely 
regard his interpretation as conjectural. Other than the appearance of 
the phrase 'asserted proposition,' there is no direct evidence to be 
found in Russell's letters to Lady Ottoline to suggest that the two men 
were discussing Moore's article or indeed any of Moore's or Russell's 
earlier views. In fact, if we are to discern anything definite on the 
basis of Russell's letters, it is that Russell was worried about whether
 the two men were discussing anything at all; what emerges from his 
reports to Lady Ottoline is not that Russell was alarmed by what 
Wittgenstein was saying but rather by whether he was saying anything.
These conversations, it must be remembered, occurred very early in 
their relationship, in fact within the first three weeks or so after 
they met. At this stage, Wittgenstein's intellectual credentials were 
not yet clear to Russell and he worries that Wittgenstein may be 'a 
fool,' 'an infliction,' and 'a crank.'[12]  McGuinness' claim that 
Wittgenstein's remark 'was clearly intended as a correction of Moore's 
position' does not take into account the serious doubts Russell had 
about Wittgenstein; it presumes that the framework of discussion between
 the two men was much more settled than appears to have been the 
case.[13]
This point is especially telling given that the position McGuinness 
attributes to Wittgenstein was, as McGuinness himself points out, 
already considered and rejected by Russell in his discussion of 
Meinong.[l4] If McGuinness is right, it is extremely puzzling how 
Wittgenstein's proposing a sophisticated view about the nature of false 
propositions and complexes which Russell had earlier considered and 
rejected could have driven Russell to suspect that Wittgenstein may have
 been, not merely wrong, but rather a fool and an infliction and a 
crank. Even if Wittgenstein had articulated his position poorly, Russell
 would presumably have (at the very least) been able to recognize the 
possibility of a position he had earlier considered.
Another serious difficulty with McGuinness's interpretation is that 
Wittgenstein states in the 'Notes on Logic' of 1913 that 'there are only
 unasserted propositions.'[15] If Wittgenstein's remarks to Russell 
about asserted propositions anticipate the opening remarks of the 
Tractatus, we must suppose that Wittgenstein changed his mind between 
1911 and 1913, and then changed it back again by the time of writing the
 Tractatus. Besides being implausible, this runs counter to a fact that 
McGuinness himself uses to support his contention that the early 
conversation anticipated ideas later expressed in the Tractatus, namely 
that Wittgenstein claimed that his fundamental ideas came to him very 
early.[16]  The continuity in Wittgenstein's thinking makes it even more
 difficult to see how Wittgenstein could have changed his mind about 
'asserted propositions' and yet have had the same ideas in 1911 and 
1918. At the very least, if McGuinness is to appeal to the continuity 
between Wittgenstein's earlier and later remarks, he owes us an account 
of the remarks from the 'Notes on Logic' concerning unasserted 
propositions.
A further difficulty with McGuinness' reading is his failure to 
offer an account of Wittgenstein's remark that 'nothing empirical is 
knowable' and how it squares with Wittgenstein's idea that only 
propositions of natural science can be said. Indeed, McGuinness argues 
that any conclusions about our knowledge that Wittgenstein drew from his
 view about the contents of the world is too 'conjectural' and cannot be
 stated 'without falling into confusion with different and more usual 
assumptions about the nature of propositions.'[17] On McGuinness' 
account, then, an important piece of the puzzle concerning that early 
conversation remains essentially unaccounted for.
Finally, on McGuinness' interpretation, remarks from the Tractatus, 
such as 'the world consists of facts, not of things' are assumed to be 
ontological claims, ontological claims anticipated by Wittgenstein in 
his conversation with Russell. McGuinness' view is that Wittgenstein was
 'correcting Moore,' both in the opening remarks of the Tractatus
and in his earlier objection to Russell. This suggests that 
Wittgenstein, Moore, and Russell shared a similar program: to offer an 
account of the furniture of the world. Where they differed, thinks 
McGuinness, was only over whether the furniture consisted of facts (or 
asserted propositions) or, concepts.[18]
However, the logical status of the opening remarks concerning the 
world and facts, and indeed the status of all the remarks of the 
Tractatus, has by no means been settled. Indeed, it is clear that for 
Wittgenstein the question 'What does the world consist of?' is in some 
sense illegitimate and nonsensical, and so too are the propositions that
 are proposed as answers to it. Moreover, Wittgenstein makes it 
abundantly clear that his aim is not to propound philosophical 
doctrines, but to show that such doctrines stem from a misunderstanding 
of the logic of the language.[19] By taking Wittgenstein to have been 
proposing ontological theses (even if these theses are seen as 
undermining all such theses) McGuinness downplays the centrality of 
Wittgenstein's antitheoretical remarks [20]
In sum, McGuinness' interpretation fails to deal adequately with 
Russell and Wittgenstein's early conversation. Not only does it fail to 
account for Russell's extreme reaction, it attributes a view to 
Wittgenstein concerning asserted propositions which is inconsistent with
 the views that he expressed shortly afterwards. As well, McGuinness 
presents very little explanation of Wittgenstein's reported remark that 
'nothing empirical is knowable' and how this squares with his idea that 
only propositions of natural science can be said. Finally, McGuinness' 
interpretation assumes that Wittgenstein's interest lies in proposing 
philosophical theories, an idea which runs counter to a fundamental 
theme of his early (and later) philosophy.
As a first step towards clarifying Wittgenstein's objection to 
Russell, it is helpful to distinguish two uses that Wittgenstein makes 
of 'assertion' in the 'Notes on Logic,' notes written within two years 
of that early conversation. In one use, Wittgenstein speaks of 
'assertion' when criticizing what he takes as Russell's confusion of the
 logical with the psychological. He says,
     Judgment, question and command are all on the same level. What 
interests logic in them is only the      unasserted proposition.
There are only unasserted propositions. Assertion is merely psychological. [21]
In this use, Wittgenstein criticizes Russell and Frege for confusing
 the psychological aspect of asserting something with the logical 
properties of a proposition. For Wittgenstein, assertion isn't a 
property of a proposition, as it is for Russell, and when we disentangle
 assertion from the real logical properties of a proposition, we are 
left only with 'unasserted propositions.' For our purposes, the 
important thing to see is that Wittgenstein's only use for 'assertion' 
in Russell's sense is critical. At this stage he would not have said 
that 'there are only asserted propositions' meaning by 'asserted 
proposition' what Russell meant by it. For that would presuppose that he
 thought that 'asserted proposition' expresses a coherent concept, 
contrary to the argument of the 'Notes on Logic.'
In his second use, Wittgenstein speaks of 'assertion' in the context
 of determining what cannot be asserted, of indicating what it would be 
meaningless to assert. Thus Wittgenstein says 'A proposition cannot 
possibly assert of itself that it is true.' He says,
     Russell's 'complexes' were to have the useful property of being
 compounded, and were to combine with      this the agreeable property 
that they could be treated like 'simples.' But this alone made them     
 unserviceable as logical types, since there would have been 
significance in asserting of a simple, that it      was complex.
As well, he declares,
     Types can never be distinguished from each other by saying (as 
is often done) that one has these but the      other has those 
properties, for this presupposes that there is a meaning in asserting 
all these properties of      both types.[22]
In the 'Notes Dictated to G. E. Moore in Norway,' written in 1914, 
Wittgenstein again uses this second sense of assertion when he speaks of
 'what is sought to be expressed by the nonsensical assertion' of 
Russell's theory of types.[23]
It is clear then that in the 'Notes on Logic' Wittgenstein thought 
Russell's notion of 'assertion' to be incoherent and that this belief is
 related to his concern with what can be meaningfully asserted, with his
 use of assertion in the second sense mentioned. If the 'Notes on Logic'
 give any clues as to what Wittgenstein might have meant in his early 
conversation with Russell, the evidence is thus against his having used 
'asserted proposition' in Russell's sense, In fact, if we stress the 
continuity of his ideas, it is likely that he would have been opposed to
 the terminology of asserted propositions in Russell's sense. For, as, 
he said, Russell's sense of assertion is psychological, despite what 
Russell himself believed, and betrays a confusion about the nature of a 
proposition. What is more likely is that Wittgenstein was using 
'assertion' in the sense of determining what counts as a meaningful 
assertion or not.
If we follow out the hypothesis that by 'assertion' Wittgenstein was
 concerned with meaning in his conversation with Russell, an interesting
 line of interpretation comes into focus. For we are able to see 
Wittgenstein's objection to Russell as questioning whether Russell's 
proposition that 'there is no rhinoceros in the room' meaningfully 
asserts anything. On this interpretation, in saying 'there is nothing in
 the world except asserted propositions,' Wittgenstein is arguing that 
Russell's proposition, that there is no rhinoceros in the room, only 
appears to assert something, but in fact does not. Since Russell's 
proposition does not assert anything, the utterance makes no sense for 
the simple reason that only propositions that assert something make 
sense. Russell's proposition about the rhinoceros would thus represent 
what Wittgenstein called in 'Notes Dictated to G. E. Moore in Norway,' a
 'nonsensical assertion,' and what he called in the Tractatus, a 
'nonsensical pseudoproposition.'[24]
In opposition to McGuinness, I am saying that the notion of an 
'asserted proposition' that Wittgenstein was employing in his 
conversation with Russell may have been radically different from what 
Moore and Russell meant by it. Far from Wittgenstein embracing Russell 
and Moore's conception of the proposition, he may have been challenging 
it on the grounds that Russell had confused nonsensical 
pseudopropositions with propositions proper. On this interpretation, he 
was challenging the very framework with which Russell and Moore pursued 
their investigations into the nature of proposition. He was not working 
within their framework and 'correcting Moore,' as McGuinness assumes, 
but aiming to undercut it.
The main difficulty for this line of interpretation is that there 
doesn't seem to be anything problematic about Russell's statement about 
the rhinoceros. 'Of course,' we want to say, 'there is no rhinoceros in 
Russell and Wittgenstein's room'; 'of course the proposition 'there is 
no rhinoceros in the room' can be asserted.' Indeed, if Russell's room 
was at all like ours, what could be a better example of a true 
proposition? How, then, can it be suggested that Wittgenstein thinks 
such a statement to be a nonsensical pseudoproposition?
Before agreeing, however, that 'there is no rhinoceros in the room' 
obviously counts as a meaningful assertion, we should pause and consider
 Wittgenstein's much later remarks in On Certainty, in which 
Wittgenstein argues that 'propositions' remarkably similar to Russell's 
proposition about the rhinoceros are nonsensical. Some examples are: 
'Here is a hand,' 'I know that there is a chair over there,' 'The earth 
existed long before one's birth,' 'I am here.' These apparent 
'assertions' are grist for Wittgenstein's mill in On Certainty, and are 
seemingly at least as commonsensical and undeniable as Russell's 
assertion that 'there is no
rhinoceros in the room.'
Part of Wittgenstein's story in On Certainty is that these so-called
 assertions, which appear to be about the way the world is, are 
assertions 'about' how we talk about the world, about the logic of the 
world, not assertions about the world at all. Moore thinks, for example,
 that he knows that he has a hand and that 'I have a hand' is an 
assertion about the world. But for Wittgenstein such a proposition is 
not an assertion, but 'stands fast' for us when we make assertions about
 the world. He says, for instance,
     I should like to say: Moore does not know what he asserts he 
knows, but it stands fast for him, as also      for me; regarding it as 
absolutely solid is part of our method of doubt and enquiry.[25]
Insofar as Moore intends us to believe that his proposition is 
something that we can know and assert, Wittgenstein regards it as 
nonsensical. In his 'misfiring attempt,' Moore is, says Wittgenstein, 
trying to describe what 'belong[s] to our frame of reference. '[26] That
 is, he is enumerating propositions that are true only in the sense that
 if they did not hold, we would lose our standards of correct judgment:
     We are interested in the fact that about certain empirical 
propositions no doubt can exist if making      judgments is to be 
possible at all. Or again: I am inclined to believe that not everything 
that has the form      of an empirical proposition is one.[27]
Interestingly, Wittgenstein harks back to the terminology of the 
Tractatus and of his earlier writings to make his point about what can 
be asserted. He says, for instance,
     My life shews that I know or am certain that there is a chair 
over there, or a door, and so on.-I tell a      friend e.g. 'Take that 
chair over there,' 'Shut the door,' etc. etc.[28]
Wittgenstein thinks it makes sense to say 'take that chair over 
there,' but that it makes no sense (at least under normal circumstances)
 to assert 'there is a chair over there.' This last assertion may appear
 commonsensical but it cannot, says Wittgenstein, be meaningfully 
asserted. For him, only the former remark can be asserted; we might say 
that, despite appearances to the contrary, 'there is a chair over 
there,' taken as a straightforward assertion, 'does not exist.'
There is a relationship, however, between so-called 'facts' such as 
'I have a hand' and my saying 'I have hurt my hand.' The former is shown
 in my saying the latter; that I have a hand is presupposed and forms 
the 'background' for the assertion 'I have hurt my hand.'[29] But 
precisely because it forms the background, it makes no sense on 
Wittgenstein's account to assert or say anything about it.
When we keep the remarks of On Certainty in mind, our initial 
assumption that 'there is no rhinoceros in the room' is a perfectly 
legitimate thing to say, begins to waiver. The issue for us, then, is 
not whether we can question the meaning of such propositions, for, as we
 have seen, propositions like Russell's are exactly the type that 
Wittgenstein spends so much time criticizing in his later philosophy. 
Rather, the question is whether Wittgenstein in his early philosophy and
 in particular in his conversation with Russell was making 
discriminations between propositions like those he makes in On 
Certainty. The question is whether Wittgenstein, in his early 
conversation with Russell, already had a sense of what counted as a 
proposition, as opposed to what belonged to the 'background' for making 
propositions. Is it plausible to think that Wittgenstein regarded 
Russell's claim that 'there is no rhinoceros in the room' as a 
'background proposition' which Russell had unwittingly assimilated to an
 everyday proposition? And is it plausible to think that in responding 
that 'there is nothing in the world except asserted propositions' 
Wittgenstein intended to point out to Russell that what Russell thought 
were propositions about the world were in fact not propositions at all?
On this interpretation, Wittgenstein was protesting that 
'propositions' about the 'background' are not propositions or assertions
 at all and that the only kind of propositions that exist are the 
everyday ones used in everyday life, propositions like those of natural 
science referred to in the Tractatus, which have 'nothing to do with 
philosophy' and are the only ones that can be said.[30] An 'asserted 
proposition' is, on this reading, something like a proposition used in 
everyday situations, something that can be said to be true or false, to 
be bipolar, to say something, to have a use. By contrast, Russell's 
statement, being about the background for asserting propositions in 
everyday situations, does not qualify as an asserted proposition since 
it cannot be said to be true or false and is not bipolar.[31]
Wittgenstein did not use the phrase 'background' in the Tractatus or
 in his earlier notes but he had the means for distinguishing sense and 
nonsense which lies behind that later idea. Specifically, his idea of 
the 'Form der Abbildung' or form of representation of a picture and 
hence of a proposition anticipates his later idea of a 'background.'[32]
 Stated in these terms, the trouble with Russell's assertion about the 
rhinoceros is that it purports to represent something that cannot be 
represented, something that belongs to the form of representation.[33] 
From Wittgenstein's point of view, Russell has 'inflated' the 
proposition about the rhinoceros and created an 'illusion of a 
perspective' in which he appears to be making a claim about the world. 
While under this illusion, Russell does not even contemplate that his 
proposition could be anything other than a straightforward everyday 
assertion.[34]
This reading of the early conversation has the advantage of 
dovetailing with Russell's report that Wittgenstein 'thinks nothing 
empirical is knowable.' If Russell took 'propositions' about the form of
 representation as empirical propositions, then he would have quite 
naturally interpreted Wittgenstein's objection as a rejection of 
empirical propositions as such. But Wittgenstein was not rejecting 
empirical propositions; he would have accepted propositions like 'the 
chair in the other room is black' as empirical and knowable. Rather, he 
was rejecting propositions that purported to be empirical propositions, 
but were not. Likewise, we can make sense of Russell's later 
recollection that Wittgenstein 'maintained that all existential 
propositions are meaningless.' Again, Wittgenstein would not have had 
any problem in accepting 'there is a black chair in the other room' as a
 meaningful existential proposition. Rather it was Russell's purported 
proposition that 'there is no rhinoceros in the room' that he considered
 to be meaningless.[35]
We can also make better sense of Russell's extreme reaction to 
Wittgenstein, and why he suspected that Wittgenstein may have been a 
fool, infliction, and crank. For Russell would have had, as indeed 
anyone who has read Wittgenstein's On Certainty is sure to have had, a 
feeling of bafflement that such apparently innocent 'propositions' as 
'here is a hand' could be viewed as objectionable. Indeed, it is 
important to emphasize how natural Russell's response is. After all, 
most of us would not have any problem thinking that 'there is no 
rhinoceros in the room' is true. All that we have to do is look about 
our room and it seems absolutely certain that there is no rhinoceros in 
it.
True, in philosophy classes we do indeed raise skeptical questions 
about such beliefs. But Russell gives no indication in his reports of 
their conversations that he and Wittgenstein were following in the 
skeptic's well-trodden path. In fact, Russell differed from Wittgenstein
 in regarding skepticism as a genuine, if mistaken, position and his 
reaction would surely have been less extreme had Wittgenstein been 
arguing a skeptical position.[38] Russell's ridiculing of Wittgenstein 
by looking underneath the desks in the room seems more connected with 
his dismissing Wittgenstein as a crank than it does with his rejecting 
an implausible skeptical argument. Moreover, we must not forget that 
Wittgenstein's objection to the rhinoceros remark was part and parcel of
 his positive contention that 'there is nothing in the world except 
asserted propositions'; this does not sound like the remark of a 
skeptic. (And remember too that, according to Russell's later anecdote, 
his objection concerned the meaning of existential propositions.) In 
short, it would seem that Wittgenstein was making a point about what can
 be meaningfully said, not about what we don't know.
It is unlikely, then, that what annoyed Russell was that 
Wittgenstein was venturing a skeptical hypothesis. What is more likely 
is that he was annoyed - to the point of suspecting that Wittgenstein 
may have been a fool, infliction, and crank - with Wittgenstein's 
actually objecting to his apparently innocent assertion that there was 
not a rhinoceros in the room. On my interpretation, Wittgenstein was 
questioning the sense of Russell's statement insofar as it pretended to 
be a species of an everyday assertion. And it is no more immediately 
obvious why there could be anything objectionable about the sense of the
 proposition about the rhinoceros than it is obvious that there is 
something objectionable about the sense of the proposition that, say, 'I
 know that I've never been to the moon.'[37]
So, if Wittgenstein's objection to Russell was indeed motivated by a
 concern with nonsense of the sort discussed in On Certainty and 
elsewhere,[38] there is a significant line of continuity between his 
views expressed in his first meetings with Russell and the very last 
days of his life. Establishing this line of continuity, however, 
requires our recognizing a much greater gap between the early 
Wittgenstein and Russell (and Frege) than is ordinarily seen. Another 
way of saying this is that if the Tractatus is to be interpreted as 
expressing a concern with nonsense, as Diamond and others have argued, 
we must be willing to take a serious second look at the development of 
these ideas in his early 'collaboration' with Russell. To this end, it 
is worth remembering what it was that Russell said Wittgenstein refused 
to admit in that earliest of conversations, namely that 'it was certain 
that there was not a rhinoceros in the room.'[39]
     NOTES
1  Information about Russell and Wittgenstein's conversation is 
derived from two main sources: Russell's letters to Lady Ottoline Morell
 and Russell's article in Mind, printed on the occasion of 
Wittgenstein's death. The first appearance of Wittgenstein is recorded 
in Russell's letter of the 18th of October, 1911, and the discussion 
about the rhinoceros appears in his letters written between the 19th of 
October and the 2nd of November.
It is interesting to note that in his article in Mind, Russell says 
that the discussion concerned a hippopotamus, not a rhinoceros. Also, 
Russell's claim to have looked underneath the desks does not appear in 
his letters to Lady Ottoline Morell. See Bertrand Russell, 'Ludwig 
Wittgenstein,' Mind 60 (1951), 297-298. Russell's letters were reprinted
 in Brian McGuinness, Wittgenstein: A Life (Berkeley and Los Angeles: 
The University of California Press, 1988), 88-89 and Ray Monk, 
Wittgenstein: The Duty of Genius (London: Vintage, 1990), 38-40. The 
quotations are from McGuinness, p. 89.
2 Wittgenstein's reported remarks about 'asserted propositions' 
occur in Russell's letters of the 7th and 13th of November, 1911. See 
Monk, p. 40.
3 My objective is to raise questions about McGuinness's hypothesis 
so as to suggest an alternative way of reading Wittgenstein's earliest 
remarks. I am not claiming to offer a definitive interpretation of that 
early conversation. As McGuinness points out, there is too little 
information for that to be possible.
4 McGuinness, p. 91.
5 Ludwig Wittgenstein, Tractatus Logico-Philosophicua, D. F. Pears 
and B. F. McGuinness (trans.) (London: Routledge & Kegan Paul, 
1961), remark 1.1.
6 Nicholas Griffin echoes McGuinness' interpretation: 'We do know 
that Wittgenstein at one point defended the views that no empirical 
propositions are knowable and that the only things that exist are 
asserted propositions. Few conclusions about Wittgenstein's philosophy 
can be drawn from these remarks, except that the second of the them is 
based on Russell's account of asserted propositions in The Principles of
 Mathematics.' See Nicholas Griffin, 'Ludwig in Fact and Fiction,' The 
Journal of the Bertrand Russell Archives 12 (1992), 79-93. The quotation
 is from p. 89.
7 It must be kept in mind that though Moore develops his realist 
conception of the proposition in opposition to Bradley's idealism, his 
version of realism is also antithetical to the conception of ideas 
derived from British Empiricism. For more on this topic see John 
Passmore, A Hundred Years of Philosophy (Harmondsworth: Penguin Books, 
1966), 202-204. For a detailed analysis of the realist reaction to 
idealism, see Peter Hylton,
Russell, Idealism and the Emergence of Analytic Philosophy (Oxford: Clarendon Press, 1990).
8 The quotation is from Paesmore, p. 203.
9 Bertrand Russell, The Principles of Mathematics (New York: W. W. Norton & Company, 1938), 49.
10 McGuinness, p. 91.
11 Ibid.
12 McGuinness, p. 89. Russell's later anecdote about the rhinoceros 
conversation sheds light on his earlier misgivings about Wittgenstein. 
He says, 'quite at first I was in doubt as to whether he was a man of 
genius or a crank, but I very soon decided in favour of the former 
alternative. Some of his early views made the decision difficult. He 
maintained, for example, at one time that all existential propositions 
are meaningless.' Note that Russell says that Wittgenstein maintained 
that 'all existential propositions are meaningless' whereas in his 
letter to Lady Ottoline at the time, he says that Wittgenstein 'thinks 
nothing empirical is knowable.' As I shall suggest, Russell's later 
remark about propositions being 'meaningless' as opposed to being 
knowable, is closer to the heart of the issue, though it is quite likely
 that in Russell's mind little rested on these different formulations. 
See McGuinness, p. 89.
13 The conversation in which, according to another of Russell's 
famous anecdotes, Wittgenstein asks Russell whether he (Wittgenstein) is
 'utterly hopeless at philosophy' and thus whether he should go into 
aeronautics or philosophy was not to take place until November 27, 1911,
 more than three weeks after the rhinoceros conversation. In response to
 Wittgenstein's question, Russell says 'I told him I didn't know but I 
thought not. I asked him to bring me something written to help me to 
judge.' It would seem that three weeks after the rhinoceros 
conversation, Russell was still having doubts. See Monk, p. 40.
14 McGuinness, p. 90.
15 Ludwig Wittgenstein, 'Notes on Logic,' Notebooks 1914-1916, 2nd 
rev. ed., G. H. von Wright and G. E. M. Anscombe (ed.) G. E. M. Anscombe
 (trans.) (Oxford: Basil Blackwell, 1961), 95. Both the 'Notes on Logic'
 and the 'Notes Dictated to G. E. Moore in Norway' are printed in 
Notebooks 1914-16. Hereafter, I shall refer simply to 'Notes on Logic' 
or 'Notes Dictated to G. E. Moore in Norway' followed by the page number
 as it occurs in Notebooks 1914-16.
16 See M. O'C. Drury, 'Conversations with Wittgenstein,' in R. Rhees
 (ed.), Recollections of Wittgenstein (Oxford: Basil Blackwell, 1984), 
158: 'my fundamental ideas came to me very early in life.'
17 McGuinness, p. 92.
18 When I insist that McGuinness has an ontological reading of the 
Tractatus, I am saying that he thinks Wittgenstein is (at least 
initially) presenting an account of the nature of language and of the 
world (and so in that broad sense is similar to Russell and Moore). I am
 aware that McGuinness differs from most interpreters in holding that 
the ultimate purpose of Wittgenstein's aims is to show the absurdity of 
all such accounts. Nevertheless, on McGuinness' interpretation, success 
in showing the absurdity of philosophical accounts of language and the 
world must rest on our understanding of the correctness of the account 
Wittgenstein initially presents. In other words, we must understand and 
be assured that Wittgenstein's (linguistic) ontology is correct before 
we can draw the consequence of ultimate 'unsayableness.' In my view, to 
interpret Wittgenstein in this manner is to admit that he has a doctrine
 after all, even if this doctrine cannot properly be said, contrary to 
Wittgenstein's disclaimer about philosophical doctrines. For more on 
McOuinneas' view,
see Brian McGuinness, 'Language and Reality in the Tractatus,' Teoria (1985), 135-144.
19 See, e.g., Tractatus, p. 3, 4.003, 4.112, and 6.53. This 
rejection of philosophical theories appears in the 'Notes on Logic' as 
well as in the 7ractatus. See p. 106 where Wittgenstein says 'In 
philosophy there are no deductions; it is purely descriptive' and 
'Philosophy gives no pictures of reality.'
20 Another way of saying this is that McGuinness has not taken 
seriously enough the question raised by Cora Diamond concerning how to 
read the Tractatus without 'chickening out.' If, as Wittgenstein says, 
his own propositions are nonsensical, it is difficult to make sense of 
how Wittgenstein can be offering an account of language and reality, 
whether that account be linguistic or otherwise. ff-Diamond is right, 
the status of the propositions of the Tractatus is altogether different 
from what McGuinness supposes. See Cora Diamond, The Realistic Spirit 
(Cambridge, Massachusetts: The M. I. T. Press, 1991), 179-204.
21 Wittgenstein, 'Notes on Logic' pp. 95 and 96.
22 'Notes on Logic,' p. 103 and pp. 100-101 (my emphasis in the case
 of both occurrences of 'asserting' and in the case of 'significance').
23 Wittgenstein, 'Notes dictated to G. E. Moore in Norway' p. 110.
24 'Notes dictated to G. E. Moore in Norway' p. 110 and Tractatua, 4.1272.
25 Ludwig Wittgenstein, On Certainty (New York: Harper & Row, 1969), paragraph 151.
26 On Certainty, paragraphs 37 & 83.
27 On Certainty, paragraph 309.
28 On Certainty, paragraph 7.
29 On Certainty, paragraph 94.
30 Tractatua, 6.53.
31 In his later philosophy, a statement about the background would 
be considered a rule of grammar. In this connection, it is interesting 
to note that in a lecture in the early 1930s, Wittgenstein discussed 
'the changes that would be required by accepting the hypothesis [i.e., 
by taking it as a rule of grammar] that there is a hippopotamus in the 
room.' Wittgenstein's point is that accepting 'there is a hippopotamus 
in our room' as a rule of grammar, as something belonging to the 
'background' would radically upset our ordinary way of seeing things; it
 would necessitate, as he says, 'queer alterations.' Though Wittgenstein
 does not explicitly mention it, it seems obvious that he thinks that we
 don't accept that proposition as a rule of grammar and that 'there is 
no hippopotamus in the room' is our accepted rule of grammar, and 
belongs to our form of representation. I might note that a 
distinguishing mark of the later Wittgenstein is that he sees 
propositions such as 'there is no hippopotamus in the room' as playing 
different roles, i.e., as rules of grammar or as empirical propositions,
 depending on the context. See Alice Ambrose (ed.), Wittgenstein's 
Lectures: Cambridge 1932-1935, (Chicago: The University of Chicago 
Press, 1982), 70.
32 The expression 'Form der Abbildung' has been translated by Pears 
and McGuinness as 'pictorial form' and by Ogden and Ramsey as 'form of 
representation.' For the purposes at hand, I do not think much 
importance rests on distinguishing these two translations. I shall use 
'form of representation' as it brings out more clearly Wittgenstein's 
interest in distinguishing the means of representation from what is 
represented. See Ludwig Wittgenstein, Tractatus Logico-Philosophicus, 
2nd rev. ed., C. K. Ogden and F. P. Ramsey (trans.) (London: Routledge 
& Megan Paul, 1933), 2.15 and 2.17.
33 We can perhaps see elements of the idea that the form of 
representation cannot be represented anticipated in Wittgenstein's often
 mentioned use of the phrase 'form of a proposition' in the 'Notes on 
Logic.' For example, he criticizes Russell for confusing the form of a 
proposition for a thing. See 'Notes on Logic' p. 105. A full discussion 
of this idea of its origins in his early philosophy would take me too 
far afield.
34  The term 'illusion of a perspective' comes from Cora Diamond and
 refers to the illusion that she thinks is created in philosophy by 
propositions which strictly speaking are nonsense. I disagree, however, 
with her interpretation that for Wittgenstein these 'nonsense 
propositions' must be understood as containing signs that have not been 
given a meaning, e.g., 'Socrates is identical' is nonsense since 
'identical' hasn't been given an adjectival meaning. In my view, 
Wittgenstein has a more robust conception of nonsense having to do with 
'uninformativenem' and misconstruing the elements of our means of 
representation. Again, it would take me too far afield to defend this 
view here. See Diamond, p. 196.
35  Some years later, when Wittgenstein had an opportunity to 
explain the Tractatus to Russell in conversation, Russell disagreed with
 Wittgenstein's view that any assertion about the world was meaningless.
 During the discussion, Russell apparently took a sheet of white paper 
and made three blobs of ink on it and asked Wittgenstein to admit that 
since there were three blobs, there must be at least three things in the
 world. According to Russell, Wittgenstein 'would admit there were three
 blobs on the page, because that was a finite assertion, but he would 
not admit that anything at all could be said about the world as a 
whole.' Russell added, 'this part of his doctrine is to my mind 
definitely mistaken.' It is possible that what Wittgenstein means by a 
'finite assertion' is similar to what I am suggesting he meant by an 
'asserted proposition' in the early conversation under discussion and it
 may very well be that this later conversation is going over terrain 
similar to that covered in the early conversation Russell's remark is 
quoted in Monk, p. 182.
36 Recall that Wittgenstein says that 'scepticism is not irrefutable, but obviously nonsensical.' See Tractatus, 6.51.
37 On Certainty, paragraph 111.
38 To say that Wittgenstein's philosophy is similarly motivated in 
this regard is not to deny the substantial differences between his two 
philosophical periods. One thing we must avoid is the fallacy that 
Wittgenstein's criticisms of the Tractatus do not contain developments 
of views first expressed in the Tractatus, though perhaps in an 
inadequate form.
39 Emphasis added. I wish to thank Paul Genest, Paul Forster, and especially Andrew Lugg for their helpful comments.