I think the future people will write English correctly

With all due respect to Fang Shonvon, it's a serious question disguised as a flippant remark. If we're going to develop a "worldwide web of knowledge", we need a lingua franca. For the present, English is a de facto lingua franca. The notion of "offshore English", while perhaps appropriate to technical ends (such as drawing up a contract, exchanging mathematical data, or defining operational goals) is most certainly NOT adapted to the *interpretation* of research.

1. THEORY-SPECIFIC and GENERALISED INTERPRETATIONS

To address the "problem of English" in discussions of the more general implications of scientific theory, we need first to address the problem – and the implications - of "generalised interpretations". Giving a **theory-specific** interpretation is a matter of assigning values to certain constants and variables within a predefined domain (or "universe of discourse"); as the term suggests, a **generalised** interpretation of a scientific theory sets the domain at its widest possible extent (the ontological realm of Frege or the early Russell, for whom – as Barry Smith remarks – "the world itself… constituted the single intended interpretation of their respective logical theories"). This extension of the domain of the interpretation raises a lot of problems: if a given hypothesis claims to account for, say, the relation between theoretical predictions of particle behaviour and observed phenomena, then its theory-specific domain covers such things as fermions, bosons, and the mathematically-modelled relations obtaining between them; however, it's an entirely different matter if we set the domain to include tables, astronomers, garden parties, intentions, fictional situations, hopes, social trends, skills, Abrahamic and non-Abrahamic conceptions of God, colours, shadows, holes, births, lives and deaths – all of which (and much, much more) come under consideration when the domain is maximally generalised.

Whether or not we hold that the theory-specific domain of particle physics subsumes all other domains depends on our preferences with respect to reductionist accounts of "the world" – and this implies, of course, taking an a priori metaphysical stance. But, whatever our position on the reductionist debate, any account we might give of, say, 'colours' or 'temporal passage' should compare and – as far as possible - reintegrate theoretical constructs with our phenomenal experience: it should show how, for example, quantum effects might underlie our phenomenal experience of colour, or to account for the explanatory gap between the relativistic account of time and our "experience" of temporality.

If I might speak frankly from my PoV as a "trained philosopher", the "Nature of Reality" board is a prime example of the problems faced by "philosophically naive" scientists when they try to draw conclusions about "reality" from scientific theory and experimental data: not only is the scientist influenced by the assumptions and intuitions about "the world" prevalent in their particular culture, but "the scientific method" is itself based on certain metaphysical and epistemological presuppositions. Save where the underlying beliefs involve logical contradictions, there's nothing "philosophically wrong" with these assumptions, intuitions, and presuppositions in themselves – what counts is THAT THEY BE RENDERED EXPLICIT IN ANY GENERALISED INTERPRETATION.

2. REALITY

While there are notable exceptions, it's usually the case that when a scientist says that such-and-such a model "represents reality", he's taking "reality" as a given. Off the record, I'd say that if one were to ask, this "given" would in most cases reduce to what we could call "a rather puzzled deictic" – reality is whatever is "out there", by some criteria of distinction between (1) what is "real" and what is "not-real" and (2) what is "internal" and what is "external" to the phenomenal subject. This is an instance of the more general problem of self-referential indexicality: when we speak of reality as being "out there", or when we make some gesture towards our environment and say "THAT is reality", we're 'pointing out' what is in fact a **private phenomenal experience**; this is a fortiori the case when our gesture is inclusive of ourselves or is given by the expression "THIS is reality".

The habit of giving "reality" by using a demonstrative – by pointing or gesturing and saying "this is real" or "that is real" – is less inexplicable if we consider that the putative referent of the gesture is not determined by correspondence between the gesture itself and "the world" but by a more holistic relation between the gesture and a certain conceptual scheme (or series of conceptual schemes). In naïve ostension, the demonstratives "this" and "that" range over a contextually-determined domain; when the domain is set to "the widest possible range" (that is, to "the whole world and everything in it"), its scope is determined by whatever cultural, metaphysical and epistemological presuppositions, assumptions, and intuitions about "the world" are prevalent in the speaker's particular linguistic and cultural group.

While we can't determine a reference to some internal phenomenal state, linguistic and cultural analyses can indeed allow us to give normative accounts of the distinctions mentioned above between what is "real" and "not-real" or between what is "internal" and "external" to the phenomenal subject: these distinctions will be largely based on the speaker's prevalent linguistic and epistemic culture, and on whatever metaphysical assumptions underlie that culture – and, whatever the problems of indeterminacy when translating from one language or culture to another, languages and epistemic systems are nonetheless **public** rather than **private**.

Of course, my choice of criteria of distinction is also influenced by my epistemic culture: distinction (1) – between what is "real" and what is "not-real" - is typical of an analytical approach to metaphysics while distinction (2) - between what is "internal" and "external" to the phenomenal subject - is usually held to be more properly the domain of phenomenology; nonetheless, both distinctions are typical of the "Western tradition" in philosophy. I have no intention of defending the Western tradition beyond pointing out its close cultural and historical relation to scientific empiricism; what is important here is that both distinctions are open to general **linguistic** investigation.

I remarked above that when a scientist says that such-and-such a model "represents reality", he's probably taking "reality" as a given. Here, we must ask WHICH *reality* he's taking as given: the "reality" determined by what is permissible to the metaphysical and epistemological presuppositions underlying the "scientific model", or the "reality" determined by his particular cultural assumptions and intuitions – or a personal mixture of the two?

3. REPRESENTATION

Not only is "reality" taken as given in the claim that "such-and-such a model represents reality": the claim also presupposes that we have a clear understanding of how anything might "represent" reality. Once again, given a theory-specific interpretation, the representation is determined by the structure of the theory itself; in physical theories, this structure is generally mathematical. At the theory-specific level, the representation is thus a mathematical model of the relations supposedly obtaining between the various elements of the predefined domain; as "reality" in such a model is set by the scope of the predefined domain, the mathematical model can be held as a "faithful representation" of its reality. However, when one sets the domain to "the widest possible scope", the local mathematical model has to be integrated at some point or another with systems of representation that are largely linguistic and (if we take the widest semiotic view) paralinguistic – without one hell of a lot of linguistic bridgework, the theory-specific mathematical model cannot give a fixed value to terms such as "table", "garden party" or "the distinction between Abrahamic and non-Abrahamic conceptions of God".

So, when we talk about "reality" in the context of scientific investigation, we're either talking about a reasonably trivial relation between the theory and its theoretically-determined domain or an immensely complex relation between the theory-bound domain of entities and relations and the universal domain of "everything we can possibly talk about". This should alert us to a very real problem: whenever we pass from the theory-specific domain to the universal (or meta-universal) domain, we're passing from a representation in which each constant and variable has, or can have, a fixed, nomologically-determined value to a representation in which the "value" of any term is determined by cultural and linguistic norms. This normativity is entirely unobjectionable - GIVEN THAT ITS RANGE IS MADE EXPLICIT. However, the closer we get to determinations based on entirely cultural criteria (such as "x is a person"), the more our understanding can suffer from indeterminacy. Imagine a language L1 in which the generally-equivalent term for "person" (let's call it "iagavag") had both a natural and a supernatural sense (much as the term "esprit" in French correlates with both the English term "mind" and the English term "spirit"). Now, any person using L1 will, according to context, use "iagavag" where an English-speaker might use "soul" and "person" – and use it correctly. However, there's a clear risk that an English-speaking native speaker of L1 might translate "iagavag" as "soul" in a context where the supernatural implication is entirely inappropriate (as a French person might speak of "spirit" when discussing the mind-body problem in English) – and there's an even greater risk that certain culturally-determined habits of thought might lead the speaker of L1 to introduce **implicit** iagavag-determined associations when using the term "person".

Furthermore, the lexical component makes up a small proportion of any language, and it would be overly reductive to limit "the representation of reality" to the putative ontological domain of a given language – we should also take into account the way the language represents relations between the different entities, and what determinant qualities it predicates of those entities. Now, if we take as an example the initial remark in this thread - "what does the future people look like?" - it's easy enough to normalise the query to "what will people look like in the future?", and I really don't think Mr Feng meant anything else. But, to turn a straw man into a simplified model, the statement "as it stands" would seem to contain a number of metaphysical assumptions that are not general in English-speaking communities. Firstly, the use of a singular verb form with the expression "the Future people" gives "people" as a class-as-one, where in English "the people" (even in the political sense) is a class-as-many and takes a plural verb form; furthermore, the use of "the people" rather than "people" could lead to the understanding that we are talking of "the people" as a political entity rather than as a group of human beings. The use of a present tense would imply that the speaker has adopted an implicitly eternalist view ("past, present, and future entities have the same ontological status"), which is certainly not the usual intuitive understanding of the English tense system. The use of "look like" reduces the domain to the physical qualities of putative future people – their morphology, colouring etc. – and not to any moral, social, or psychological qualities (as would be the case if the query were "what will future people be like?").

Now, all this might sound like typical linguistic nit-picking, but when we're dealing with philosophical questions, language is important – after all, it's the ONLY tool we have for doing philosophy. When we're dealing with big, nebulous concepts like "the domain of all there is" or "the representation of reality" we can't just take them as "given" - any term or structure we use is laden with culturally-determined associations and assumptions. Any generalisation we might make from a theory-specific interpretation passes from a clearly-defined domain and a syntactically-explicit system of representation to a domain and a system or representation, which is normative rather than nomological. This is as true for a native English-speaker as it is for a non-native speaker; however, the English-speaker has more implicit awareness of the normative range of the terms she employs than would the non-native speaker – a situation which cuts both ways in cross-linguistic communication, as the English-speaker can easily employ terms in a normative sense that is not available to her interlocutor

SO?

This message isn't just an exercise in linguistic one-upmanship – it has two clear objectives. The first is to point out that having technical competence in English within a given area of investigation doesn't guarantee that we can make the transition to using English when discussing the wider ontological and metaphysical implications of the area; if this were my sole objective, I would argue for the extension of "offshore English" to include a better understanding of the grammatical and lexical categories of English - that is, backing up "efficient language use" not only with a minimal understanding of "what the structure implies" about the way English represents the various entities, relations, and properties that make up the world, but also with a comparative understanding of how the English representations parallel representations in the native language. However, my remarks here can be generalised to any discussion of "what scientific theory tells us about reality", even when the discussion is monolinguistic; given the de facto role of English in scientific debate, I shall restrain my final remarks to English as a natural language.

When we pass from a theory-specific interpretation to a generalised interpretation, we pass from an interpretation, which conforms, to a formal or semi-formal language to an interpretation, which employs the normative resources of natural language. The representational mode of the theory-specific interpretation is determined by the formulation of the theory – as I remarked above, in physical theory this is generally mathematical; other sciences employ natural-language terms with a fixed, explicit sense. However, when we pass into natural language, our structure is no longer explicit and our terms are no longer fixed. Thus, a term with a reasonably well-defined sense in a given area of investigation can be the source of confusion and misunderstanding when used to relate the area of investigation to a more general understanding of the world – the term "event" is an excellent illustration. In the first chapter of his introductory text "General Relativity from A to B", Robert Geroch characterises the notion of "event" as follows

"The notion of an event is the basic building block of the theory. … By an event we mean an idealized occurrence in the physical world having extension in neither space nor time. For example, "the explosion of a firecracker" or "the snapping of one's fingers" would represent an event. (By contrast, "a particle" would not represent an event, as it has "extension in time"; "a long piece of rope" has "extension in space"). By "occurrence in the physical world" we mean that an event is to be regarded as a part of the world in which we live, not as a construct in some theory"

Now, Geroch is most certainly positioning his explanation in the "widest possible domain" ("an event is to be regarded as a part of the world in which we live, not as a construct in some theory"). He characterises events as "idealised occurrences" in the physical world "having extension in neither space nor time", then gives as examples events that are extended in both space and time. From the basis of this characterisation, he disregards the possibility that a particle could be an event (it has "extension in time") – yet given that both "the explosion of a firecracker" and " the snapping of one's fingers" effectively DO have extension in time, there is no a priori reason for counting them as events while counting particles as not being events. Once again, we might argue that this is "not what Geroch meant" – it most certainly is not (and, to his credit, he extricates himself a bit in his further development). But it most certainly is what Geroch is actually saying, and his choice of analogy illustrates his underlying confusion about what events and material objects might "really" be. Linguistically, he's confusing the description of a punctual event with an event described punctually; metaphysically, he's confusing the notion of an illimitably short interval with that of an instant. Furthermore, he's taking for granted we fully understand what events can be regarded as "a part of the world in which we live", and that we understand it as fully as we understand "a construct in some theory". Without going into the question of whether the "events" described by relativity are theoretical constructs or not (they are certainly geometrical constructs), the problem with Geroch's characterisation has nothing to do with his grasp of relativity, but rather with his grasp of the representational powers of natural language. The distinction between, for example, physical entities and events can't be reduced to a simple distinction either of temporal duration or of physical extension, but is rather a matter of how we differentiate between the temporal duration and physical extension of each class. Furthermore, neither class is given in extension, so we don't know whether his characterisation of events is supposed to cover such disparate things as explosions, geological events, parties, changes, voyages, beginnings, or deaths – and of this range, only changes (and, by extension, beginnings) are even susceptible to description as "instantaneous" in the sense targeted by Geroch.

These observations concern questions of ontology – "what kinds of thing constitute the furniture of the world, how do we differentiate between them, and do any of them constitute the ultimate furniture of the world?". When applied to a theory-specific interpretation, such questions are answered by the giving the domain fixed in the theory itself; when applied to a generalised interpretation, we need either to show very clearly at what point a theoretical term – such as "event" – touches the equivalent natural language term, and how the range of the natural language term diverges from that of the theoretical term. Of course, once we've determined possible answers for the "how", we have to imagine possible answers for the "why" – if large-scale, long-duration occurrences such as geological events are not "events" in the sense of relativity, what are they – and what relation, if any, is there between "geological" and "relativistic" events?

We can make similar remarks about the more syntactic and structural elements of an interpretation. While it employs certain conventions that are common with natural language, the syntactical structure of a theory is fixed and invariable, while that of a natural language is largely open to contextual determination (compare the logical connector in "he died & he went to Paris" with the natural language conjunction in "he died and he went to Paris"). In natural language, sense can determine structure in the same way that structure determines sense. All in all, I'm tempted to say that a very large proportion of our "intuitive metaphysical assumptions" are determined by the resources of whatever natural language we employ – it's certainly the case when dealing with things like duration, persistence, cross-temporal relations, or frequency; I'd also hold that a substantial proportion of such assumptions are rather normative than nomological.

If my remarks are justifiable (and, although I'll reserve further development for the moment, I believe they are), then we should be wary of any passage from the fixed ontologies, quantities, and relations of a theory-specific interpretation to big, nebulous concepts like "the domain of all there is" or "the representation of reality". As soon as we attempt to integrate the theory-specific interpretation with our wider understanding of "the world", we must examine as precisely as possible not only our theory but also what constitutes our "wider understanding of the world". A possible role for philosophy in all this is (paraphrasing Craig Callender) to "refine our description of what needs to be explained, carefully examine science and the way it treats the phenomenon under consideration, compare the two, and then try to account for any explanatory gap that arises. The gap may be filled in with scientific or philosophical resources". Here, my interest is the "refinement of our description" and the "philosophical resources" that such refinement might require – resources than can only be ignored at his peril by the non-philosophical scientist. And, foremost among such resources is the realisation that our ontologies are determined by our language, and by our pragmatic ends in using language in a certain way in a given situation – an observation that applies as fully to theory-specific interpretations as it does to the vaguer, more normative "ontologies" we can suggest for natural languages.

Not only does contemporary philosophical investigation employ powerful tools which allow closer examination of ontological and metaphysical questions (mereology, non-standard logics, and possible world analysis being good examples) - it has also developed a tradition of examining puzzling notions such as "substance", "body", "property", "relation", "identity", or "individuation", notions which are fundamental to any understanding of our everyday talk about things as individuals, masses, causal relata and the like. The primary aim of philosophy is not to talk about the world, but to examine the WAY we talk about the world – and as such, philosophy represents an **extension** of our ways of talking about the world: Quine notably considered philosophy to be a "continuation" of natural science; the project suggested by Callender sees philosophy as helping in the determination of a wider, collective understanding of the world which integrates, as far as possible, different epistemic systems.

I worked it up to 46 pages, dropped the comments about Geroch, and ended up suggesting a new approach to the analysis of professional organisations... funny thing, research.

I've just finished a paper on what we can, and cannot, accept in our "widest understanding of the world". I'm afraid that the scientific account is one of, but not the sole, account I take into cosideration.

Brad Pitt and Angelina Jolie, I hope.