Re: Popular(izing) Science

From: Damien Broderick (d.broderick@english.unimelb.edu.au)
Date: Fri Feb 09 2001 - 05:12:33 MST


At 10:35 PM 7/02/01 -0500, Josh Martin wrote:

>I very much want to be a
>popularizer, to attract fresh blood to the science, as well as to inform
>John Q. Public about the wonders of science, and what it can do (and has
>done) for his life.
>
>That said:
>How can I get there from here? (ATTN: Mr. Broderick and other
>writers/communicators on this list)

Josh, I think it's primarily a matter of speaking/writing as clearly and
entertainingly as possible, without dumbing down so far that the content is
abused. But the fundamental cultural problem is deeper and more wounding
than the need for simplified Asimovian storytelling. I don't know if what
follows helps in isolation from the rest of my argument (in the book THE
ARCHITECTURE OF BABEL), but I throw it in here for what it's worth, even
though it's pitched at a trained audience. Unfortunately, the indents and
other formats are lost in email:

===================================

Despite superb attempts by specialists such as Paul Davies and John Gribbin
to turn chaos and quantum theory, cosmology, genetics and ecology into
popular readable form, the strict working discourses of the sciences are
quite simply incommunicable to the non-initiate. Today's poetry might be
hard to understand (though not, surely, as thorny as the most formidable
literary criticism), but harder still is real cutting-edge science. The
reason is sketched by Kuhn:

                The most esoteric of poets. . . is far more concerned than the scientist
with lay approbation of his creative work, though he may be even less
concerned with approbation in general. That difference proves
consequential. Just because he is working only for an audience of
colleagues, an audience that shares his own values and beliefs, the
scientist can take a single set of standards for granted. (Kuhn:164)

        Even in essays at popularisation, science rapidly evades the untrained
understanding. The non-specialist is unlikely to relish the following
passage, I think, any more than the average mathematician would enjoy a
paragraph from Derrida--yet Philip J. Davis and Reuben Hersh's The
Mathematical Experience is meant to be a popularisation!

                Although the nonstandard universe is conceptually distinct from the
standard one, it is desirable to think of it as an enlargement of the
standard universe. Since R* is a model for [the formal language] L, every
true sentence about R has an interpretation in R*. . . .
                An object in R* (a pseudoreal number) is called infinite if it is
pseudogreater than every standard real number; otherwise it is called
finite. . . . If the pseudodifference of two pseudoreals is finite, we say
they belong to the same `galaxy'; the pseudoreal axis contains an
uncountable infinity of galaxies. If the pseudodifference of two
pseudoreals is infinitesimal, we say they belong to the same `monad'. . . .
The notion of a monad turns out to be applicable not only to real numbers
but also to general metric and topological space. Nonstandard analysis
therefore is relevant not just to elementary calculus but to the entire
range of modern abstract analysis. (Davis and Hersh, 1981:251-2)

        Am I playing fair here? Perhaps that book is not truly a popularisation.
Perhaps its market is really those readers who once did quite a lot of
mathematics but have let its finer details slip from memory. Well, then,
consider the following instance, drawn from a tribute in Scientific
American to the late Richard Feynman:

                [Feynman's] lecture is a tour de force of exposition, seeking to derive a
couple of the most important results of quantum field theory not `in the
spirit of Dirac with lots of symbols and operators' but by explicit
arguments that flow from those summary zigzag sketches known everywhere as
Feynman diagrams. True to his way Feynman explains that he will set out
some `very simple examples. . . because if you do you will understand the
generalities at once--that's the way I understand these things anyhow'
(`Science and the Citizen', 1988:25. Author's ellipses.)

        Is it true that the `summary zigzag sketches' are known everywhere?
Hardly. The terms `DNA', `entropy' and `chaos', and a little of their
denotations, might have become educated commonplaces in the last two
decades, alongside certain computer terms, but knowledge of Feynman
diagrams is perhaps largely restricted to those non-scientists with a taste
for high-energy mysticism (e.g. Zukav, 1979:233-61) and the comparatively
few readers of journals such as Scientific American. The interesting step,
however, follows in this `popularisation':

                The diagrams are a form of shorthand analysis, but they are far from a
geometrization of the events. Their fidelity relies on careful and original
mathematical rules that accompany each straight or wiggly line, usually
generating at every vertex an exercise in matrix algebra; between
intersections they imply a clever nesting of integrals that serve to sum
over the spacetime excursions that intervene. (Idem)

        Feynman's sorrowing colleague is not being sardonic here, of all places,
at the expense of his non-specialist readers. Nevertheless, the deployment
of this precise (though highly general) account of Feynman's achievement
takes on the aspect, to the lay reader, of liturgy: of grand rhetoric,
poetic because we trust--on faith--its puissance to the initiate.
Otherwise, it is sheer mumbo-jumbo.
        I once made this comment to a science journalist, and she was mightily
offended. Feynman's theories on quantum electrodynamics (QED for short)
were by no means impossible for the general reader to understand, she told
me--for the unusual reason that Feynman himself wrote one of the greatest
science popularisations on that very topic. His book, based on a series of
lectures for a lay audience, is called QED (Feynman, 1990), a neat little
joke in itself. (For those who have forgotten their geometry lessons,
Q.E.D. also stands for quod erat demonstrandum, the phrase used to sum up a
demonstration.) It is lively, avoids pomposity and tricks of higher
mathematics--and, in my view, is really a snow job. I do not think it
explains those wiggly lines at all. It tries very hard, but in the end its
readers will carry off nothing but a blurry sense of a wonderful truth
forever out of focus.
        I feel confident about this, because Feynman himself was honest enough to
admit it in the book. `The. . . reason that you might think you do not
understand what I am telling you is, while I am describing how Nature
works, you won't understand why Nature works that way. But you see, nobody
understands that' (ibid.:10). True, this is a subtle distinction. Perhaps
the how can be conveyed, even if the why is forever out of bounds. Not so;
in another typical friendly joke, the great physicist opened his third
lecture this way:
        
                Some of you haven't heard the other two lectures and will find this
lecture almost incomprehensible. Those of you who have heard the other two
lectures will also find this lecture incomprehensible, but you know that
that's all right: as I explained in the first lecture, the way we have to
describe Nature is generally incomprehensible to us. (ibid.:77)
        
        Well, perhaps this is still just a humble acknowledgment that our poor
human minds can never hope to plumb the cosmos to its bottommost depths.
When we get into Feynman's own special mastery, though, he is forced to
shrug his shoulders in despair:
        
                The second action fundamental to quantum electrodynamics is: An electron
goes from point A to point B in space-time. . . . The formula for the
amplitude of this action. . . is a rather complicated formula, and I'm
sorry that I don't know how to explain it in simple terms. (ibid.:90-1)

        We do not need to get into the amplitude of quantum least-action paths to
get lost. What can the curious onlooker from the humanities make of a
General Motors advertisement in the front of the same issue of Scientific
American that carried Feynman's obit? (Remember, what follows is from a
paid advertisement in a popular magazine!) Expensive, handsomely mounted
over three pages of colour for its impact on sales and perhaps staff
recruitment, the ad describes research into robotic arm control. It is
worth citing about a fifth of the text, for the accumulating impact it
achieves on the dazed, utterly excluded humanities reader:

                If the problem could be discretized, making it in some sense finite, it
could be put on a computer and solved numerically. So Dr. Marin replaced
the unknown function with a piecewise cubic approximation.
                This allows the search for the unknown function to be confined to a class
of functions that are completely characterized by a finite number of
coefficients in a B-spline series.
                He similarly discretized the constraints, replacing the infinite set of
constraints with a finite dimensional subset that could be dealt with
numerically.
                He completed the formulation of the discrete problem by incorporating a
grid-refinement strategy. Now the problem's dimension could be gradually
increased to better approximate the continuous case.
                What resulted was a classic non-linear optimization problem, a finite
dimensional problem in which it remained only to find the coefficients of
the B-splines while satisfying the constraints.
                A monotonicity property of this problem coupled with properties of the
approximation method suggests that the simple technique of cyclic
coordinate descent might best provide a solution. (Scientific American,
June, 1988:17)

        The shape of the language is cheerful journalese. It goes on at once to
quote from Dr Marin, and ends with biographical notes (`Sam is a Member of
the Society for Industrial and Applied Mathematics. He lives in Rochester
Hills, Michigan, with his wife and two children.'). Nor is it difficult to
decode the general sense of the text. It is hardly more technically obtuse,
on the face of it, than any carpenters' guide to fixing the rafters, which
is also a complex discourse laden with specialised terms that exclude the
outsider but make for precision and speed among practitioners. The problem
is, however, one of kind, and not of degree. Consulting a dictionary may
provide definitions of `monotonicity' but this will be hardly enough to
make it operational. The Britannica explains without enlightenment that

                In problems in which the approximate has to match f(x) but not Df(x),
D^2f(x), etc., it is undesirable to introduce derivatives of f(x) as
additional parameters (arbitrary constants). Consequently, a desirable
property in piecewise functions might be continuity of the derivatives at
the nodes without specifying the values of the derivatives there. Piecewise
approximates of this type are known as splines, the most popular form of
which is the cubic spline. (Encyclopedia Britannica, Vol. 25:43)

        `Elucidation' of this order points to an exclusion principle at work: the
two cultures remain sundered.
        By contrast, we might consider a fragment from Wallace Stevens, premier
poetic object of American deconstructive analysis (cited and discussed in
Paulson, 1988:102):

                Tell X that speech is not dirty silence
                Clarified. It is silence made still dirtier.

        This takes the form of an injunctive proposition. It looks like a
philosophical assertion rendered rhetorically for greater bite. It might
even be supposed to enact itself, for it is at the least emulated speech
made dirtier, in the sense of more opaque, quarrelsome, difficult. Less
translucently, it might infold or hide profound intertextual nuance and
ambience. Perhaps to take its measure fully one needs to be acquainted with
all Stevens' work, with that of his contemporary poets and perhaps
philosophers, and with the cultural matrix out of which his reading and
writing was ushered. Charles Bazerman makes this point in an
interdisciplinary examination of `what written knowledge does', though he
is dealing chiefly with critical discourse in literature and `first order'
discourse in science. Of the former he observes:

                The literary vocabulary on one level appears to be purely technical, not
unlike the technical vocabularies of molecular biology or sociology. Terms
such as topos, apostrophe, sonnet, turn, enjambment and sublime are the
critic's basic conceptual equipment, learned as part of professional
training. On another level, however, the literary terms are more than
technical, for each reverberates with former uses and examples. One can
know and understand deoxyribose on the basis of modern chemistry alone, but
to understand the sublime one must not only have read Longinus and be
familiar with the ensuing critical debate to modern times, one must have
experienced a wide range of poems that embody the development and variation
of that concept. (Bazerman, 1981:376)

        Still, surely the lay reader of literature (and even of much criticism)
has a grasp of its meaning which vanishes when faced with B-splines and
cyclic coordinate descent-based algorithms. The paradoxical outcome of this
fact is drawn by William Paulson, who cited the Stevens poem in
exemplification of a rueful remark by the great physicist Paul Dirac, `In
physics, we try to say things that no one knew before in a way that
everyone can understand, whereas in poetry. . .' (cited in Paulson,
1988:101). While it might not be true that in poetry the everyday is
rendered as obscurely as possible (is it really mere common-sense that
speech is cleaned-up noise?), a cultivation of plurivocity, aporia and
level-shifts of meaning is patently and typically signalled by Stevens in a
way absolutely unlike the attempt on precise disambiguation in the GM
advertisement and the Feynman obit.
        If the poet dirties the windowpanes of silence and speech alike, it is `to
make the visible a little hard/ To see' (idem): to interrupt utility in the
interests of insight--even if that insight is always-already `blind', as
the late American deconstructor Paul de Man would have it (de Man, 1971).
Paradigm challenge, on a microlevel, is being raised to a principle of
studied textual generation. But note that this claim of mine itself
partakes of both methods, aiming at explanation while unashamedly
`bisociating' two meanings of paradigm: the Kuhnian, where the word means
an exemplary model practice, and the literary, where it means a list of
synonyms. We cannot easily, if at all, dispel ambiguity from discourse.

Signals out of Noise

I began with the sociological observation that if any unified culture had
ever existed it is now lost to us. With very rare exceptions, trained minds
and sensibilities within the western episteme may be cultivated in an
artistic or a scientific mode but not both. Snow's aperçu captured this
truth of the industrial and postindustrial condition with journalistic
vividness, simplifying the great grinding tectonics of history, politics,
knowledge-constitution, power into a slogan: `the two cultures'.
        It has never approached adequacy as an analytic account, nor, I think, was
it so intended. Snow meant his rude outburst to shake a complacent
establishment into heeding the advice of those `men of affairs' whose
scientific specialties lent them a privileged access to the world's
abundant human miseries, and to a historically new class of actual or
potential remedies for those disorders. Today we more easily read his
slogan in terms of competing discourses. Not just two, of course, for
discursive theorists have shown us how multiple and proliferative these
communicative practices are. Still, for polemical purposes two might be
after all the optimum categorial number.
        Cliche runs on ahead of any discussion. The humanities and the sciences:
the poetic, subjective utterance of the creating artist, versus the
stringent, objective notation of the observing scientist. Neither summary,
as we have seen, is remotely competent, yet both retain a powerful hold on
the imagination of our epoch. This is no accident, nor is it purely a
function of ideology.
        Critical theory at the end of the millennium embraces aporia and rupture.
Is it, then, deeply wrong-headed even to ask whether some alternative
discourse might be able to bridge these two isolated empires? Whether it is
speculative narrative, postmodern collage, the postfuturist sublime or some
other candidate, perhaps the enterprise is intrinsically misconceived?
Perhaps what Snow witnessed was no less than the first unmistakable rumble
of what would become a tumultuous geyser as the stressed epistemic systems
buckled, tore asunder, burst open under contradiction like some allegory of
dialectics.
        But perhaps this account is too apocalyptic by half. Even if it is the
case that our episteme is undergoing crisis (assuming that all the
multitudes of vectors in a culture can be summed into a single arrow), we
have no warrant either to wring our hands or to rejoice at this complex
fact. All that can be said with certainty, perhaps, is that our language is
always-already inadequate to inscribe prescriptions or memorials on such a
grand scale.
        There is no outside-the-text, because everything which a multi-levelled
subjectivity uses to turn graphemes into meaning is `inscribed' in some
text or other, some semiotic system. Even our apparently direct access to
the world arrives through narrow sensory codes and channels and vast
quantities of neural pre-processing.
        There is no inside-the-text, because a text is just an array of abandoned
lexemes until it is activated by semiotic systems of textual codes, frames
and encyclopaedias operating inside and around its readers.
        The literary theorist Douwe Fokkema, as we have seen, posits three textual
supercodes: the syntactic, the semantic, and the pragmatic. Any literary
act of writing or reading, on this analysis, invokes other subsidiary
codes, such as the group- or period-codes which constitute and signal one
level of inscription and reception. But so too does any scientific act of
notation, computation, reportage, propaganda, textbook dissemination of the
canonical and the paradigmatic.
        Written communication is hardly all there is to scientific discourse,
despite Latour's emphasis on inscription, for much of science is conducted
by tacit means. Through apprenticeship, techniques and protocols of
interaction are learned and transmitted by non-formal and non-textual means
(unless `textual' means, vacuously, everything). Nor is textuality
everything in painting, music, sculpture, acting, film-making, or perhaps
even writing and speaking poetry and prose. Language is the chief part of
what we can pin down and analyse in these practices. What most strikingly
segregates the literary and the scientific communities is their separate
patterns and approved protocols of discourse, of communication and
publication. Still, we must not forget, as a cognitive philosopher reminds
us, that there is `something important missing from the sociological
account. That something is. . . causal interaction between scientists and
the world' (Giere, 1988:4). Charles Bazerman expresses well the crux of
this impasse:

                Our current inability to forge a convincing link between the socially
constructivist critique of scientific formulations and the empirical
project has roots in how we have become accustomed to think about language
in this century. When socially minded observers of scientific activity come
to think about the role of language, our current concepts of language offer
no strong clues about how language talks about anything other than itself.
. . .
                [D]ebate over how to talk about one's subject continues in all
disciplines today, and cannot be separated from the fundamental practices
of those disciplines. [. . .I]n those communal endeavors whose goal is
symbolic knowledge, the more we understand the way symbols are used in the
activity, the better we can carry out that activity. (Bazerman, 1988:295, 317)

         At the same time, it is certainly not aporia and paradox versus precision
and univocity that serve to distinguish the humanities from the sciences.
As we have seen, both hegemonic systems engage regularity and discontinuity
by turns. This is why Jean-François Lyotard stresses the importance in
science, as in the arts, of paralogy: of what appear to be errors or wrong
moves in a given game, but prove to be superior moves (or at least to have
the possibility of being so) in a slightly or a drastically new game
(Lyotard, 1984).
        Philosopher of science Paul Feyerabend, an `epistemological anarchist',
has argued that science and the humanities pursue very much the same mixed
strategies and tactics. A scientist, for example,

                starts with a bulk of material consisting of diverse and conflicting
ingredients. There are theories formulated in accordance with the highest
standards of rigour and precision side by side with unfounded and sloppy
approximations, there are `solid' facts, local laws based on some of these
facts, there are heuristic principles, tentative formulations of new points
of view which partly agree, partly conflict with the accepted facts, there
are vague philosophical ideas, standards of rationality and procedures that
conflict with these. Being unable to make such material conform to simple
views of order and consistency the scientist usually develops a practical
logic that permits him to get results amidst chaos and incoherence
(Feyerabend, 1978:199).

        William Paulson has urged a similar view--abhorrent to traditionalists in
either culture--with force and zest. In the light of our current
understanding, he notes, it is reasonable to seek in literary studies (and,
one might add, paraliterary and scientific studies)

                not pure knowledge, not a means of dishing out common cultural fare to
the many, but a part of the process wherein new concepts, practices, and
symbolic moves are invented and propagated. . . .
                Literature is the noise of culture, the rich and indeterminate margin
into which messages are sent off, never to return the same, in which
signals are received not quite like anything emitted. (Paulson, 1988:180)

        The frame for Paulson's startling argument is the place we started our
discussion: information theory understood within a model of life and
culture as self-organising, creating their own stability and complexity in
the face of entropy, a picture based on Prigogine's controversial theory of
thermodynamics far-from-equilibrium (Prigogine and Stengers, 1984; Waldrop,
1993; Lewin, 1993). Noise is not necessarily just a `parasite' (as the
French call it, in an untranslatable pun). It can be novel information that
requires a new frame of reference for its reception. It can be an
unsettling input into a tired old system from one or more alternative
structures. Whether smuggled or welcomed into that system, redefining it,
such novelties can renew and extend its possible moves in the world--can
renew even its self-understanding. Paulson notes:

                Literature is a discursive form and an archive that is out of step with
technocratic modernity. It cannot therefore supply that modernity and its
political institutions with reassuring narratives or metanarratives to live
by; it can only perturb them as a kind of noise. Yet it is only because of
internal perturbation, whether arising from memory or from the plurality of
cognitive and discursive strategies, that we can be observers of ourselves.
. . . Internal noise, in other words, keeps us from being so fully
integrated into a silently functioning system that we would cease to be
aware of it as system. (Paulson:181)

        But while this is true of canonised literature, alternative textualities
such as postmodern, post-colonial and other marginalised narratives might
have a larger role still. Unlike authorised literature, these texts
sometimes have the capacity to grind noise and grit into the
meta-narratives which ideology remorselessly drones through our eyes, ears
and bodies. By a curious reversal, they constitute a discursive form and an
archive which is not always out of step with technocratic modernity or
postmodernity. True, this might make such textuality easily co-optable to
the work of suasion; the postmodern workplace, including the new virtual
reality cyberspace, is more often decorative than challenging (Benedikt,
1991; Rheingold, 1991; Ross, 1991). Yet it offers as well an enhanced
access to Paulson's fecund noise.
        For most of this century, since the triumphant merger between genetics and
Darwinism, we have known beyond doubt that animals, plants and microbes are
machines for understanding the world. Calling living creatures `machines'
still disturbs many people, because traditionally machines are instruments
made by humans for human purposes. The natural world is not a table set for
our hungry and feckless consumption. Agreed; but on this scientific account
we, too, are machines--wonderful, non-linear machines with feelings,
aspirations, duties and a hunger to know and do.
        Humans, like all organic machines, have been shaped by evolution to deal
with their environment, and our principal environment is other human
beings. So we have specialised in comprehending one another: face to face,
in groups, through all the media of discourse (Schank and Abelson, 1977).
Most of the time our brains are dedicated to constructing stories about
people. The human brain/mind is primarily a hermeneutic device tailored for
interpreting the actions and motives of others.
        Emulating the messy, vastly cross-linked neural architecture of the brain,
distributed parallel-processing computers are now starting to model how we
can understand, intend and act almost instantly in a complicated world
(Tank and Hopfield, 1987:62-70). If people make plentiful mistakes of
interpretation, that is the price we pay for functioning in a real world
indefinitely more complex and noisy than the awful neatness of the
traditionally scientific linear programs driving orthodox computers. Minds
are `knowledge systems', not `logic systems', operating what Jeremy
Campbell has dubbed, in a memorable phrase, the worldly brain (Campbell,
1989).
        Where a computer needs to be told by an explicit algorithm to go to a
specific `address', a pigeon-hole in memory that might contain anything, a
human mind reaches directly for a desired `content'. This information
(including codes for feelings, wishes and the rest of what makes us human)
is dispersed throughout the brain, each location linked to others by
networks of affinities. One idea usually leads to another via schemata,
standardised stories we tell ourselves about how our world works. And in
most cases, what we need to know about that world is how other people
think, feel, expect and mean to act. The arts and humanities, one might
say, are the first archive and continuing matrix of practices born from the
evolved machinery which defines us as Homo sapiens sapiens.
        The human world that our worldly brain is dedicated to understanding is,
then, in a profound sense far more complicated than the world at large,
which is why science is paradoxically so simple. The practice of science is
obscure because it manipulates esoteric but starkly elementary models very
removed from the domain we inhabit (even though its results are absolutely
pertinent to everyday life). `The kind of beauty we find in physical
theories,' notes Steven Weinberg,

                is of a very limited sort. It is. . . the beauty of simplicity and
inevitability. . . Shakespeare's plays. . . are big messy compositions
whose messiness mirrors the complexity of life. That is part of the beauty
of his plays, a beauty that to my taste is of a higher order than the
beauty of. . . general relativity' (Weinberg, 1993:119)

        One surprising consequence of such demythologised accounts is a
realisation that both great systems of discourse, despite Weinberg's
gallant bow of deference to Shakespeare, have a lot in common. For many
scientific researchers hard at work decoding the objective world,
Feyerabend reminds us,

                metaphysical principles are used to advance research, logical laws and
methodological standards are suspended without much ado as constituting
undue restrictions, adventurous and `irrational' conceptions abound. The
successful researcher frequently is a literate man, he knows many tricks,
ideas, ways of speaking, he is familiar with details of history and
abstractions of cosmology, he can combine fragments of widely differing
points of view and quickly switch from one framework to another. He is not
tied to any particular language for he may speak the language of fact and
the language of fairytale side by side and mix them up in the most
unexpected ways. (Feyerabend, 1978:200)

        Perhaps it is in this odd but charming rapprochement that serious inquiry
into the sciences and humanities must finally discover itself: in the
underdetermination of all theory, evaded by the defiantly or unknown
uninscribable, and in its overdetermination by a multitude of factors
beyond computable assay; in the discourses of a worldly brain dedicated to
the understanding of others like itself, yet dwelling in a world rebuilt by
its own ingenuity. For that very reason--the ubiquity and at last the
banality of the radical `incompletability' of knowledge--one is obliged
(like every scientist, like every poet) to continue dipping the sea with a
sieve.

===================================

-------------------------------------------------------------------------
Dr Damien Broderick / Senior Fellow, Department of English and Cultural
Studies
        University of Melbourne, Parkville 3052, AUSTRALIA
        @: <d.broderick@english.unimelb.edu.au>
        Unofficial biography/bibliography site:
        http://www.thespike.addr.com
Australian Business Number (ABN): 98 674 378 153
-------------------------------------------------------------------------



This archive was generated by hypermail 2b30 : Mon May 28 2001 - 09:56:38 MDT