The significance of the LANL preprint server

From: Ransdell, Joseph M. <ransdell_at_DOOR.NET>
Date: Sun, 11 Jul 1999 20:57:22 -0500

The magazine about academia called "Lingua Franca" has a page in the
most recent on-line issue listing what it calls "The Tech 20", meaning
people who seem to be especially significant for some reason or other in
connection with "the digital revolution" (in the opinion of Jens David
Ohlin, who apparently composed that page). Paul Ginsparg is among them
and the blurb on him runs as follows:

> OCCUPATION: Physicist the Los Alamos National Laboratory
>
> WHAT HE'S DONE: Do too many peer reviewers spoil a scientific
> paper? Not according to Ginsparg—or the many physicists who
> visit his Web site (xxx.lanl.gov). Ginsparg has created an on-line
> database of scientific papers, available to the entire professional
> community for reading and commentary before their publication
> in print journals. Since its genesis in 1991, the Ginsparg Net (as
> it is now known) has grown from a resource serving two hundred
> physicists to an international archive accessed by more than sixty
> thousand users in eighty countries. Scientists in the developing
> world enjoy the same level of access as a graduate student at
> Harvard.
>
> WHY IT MATTERS: Instead of waiting for results to be published
> in print journals with months-long publication lags, physicists
> now get immediate access to each other's research findings.
> Physicists who post papers receive private emails from those
> choose to critique their methodology and results. The writer
> then resubmits a revised draft, which is posted on Ginsparg's
> site. A potentially unlimited number of scientists can
> contribute to the development of each author's work
> —strengthening the final article before it goes to a print
> journal.
>
> QUOTE: "The essential question at this point is not whether
> the scientific research literature will migrate to fully electronic
> dissemination but rather how quickly this transition will take
> place now that all of the requisite tools are available."

http://www.linguafranca.com/9907/tech20.html

This is an interesting interpretation of the significance of the LANL
preprint system, and there is some truth in it, but there is something
that is obscured in thinking of it this way, too, and clarity on this is
becoming more and more important as things move along in the archiving
initiative Stevan Harnad is leading. I say this in part because I am
bothered by the fact that the significance of the LANL archive as a
model inspiring the present movement increasingly appears to have little
or nothing to do with its function as a preprint server even though it
was nothing but that until recently! But now it seems that it is really
the overlay functions which put it in the service of the refereed
journals, plus the addition of various librarial hypertext features that
counts. Did I have it wrong to start with in thinking the preprint
server important? I don't think so, and what follows is intended to
explain why.

I remarked earlier that the Caltech Proposal seemed not to have the Los
Alamos system in proper perspective, as was shown in there being nothing
in the system proposed that corresponds operationally to the preprint
server -- which, however, did not even stimulate an attempt to explain
the seeming discrepancy away. Since then I have noticed also the
conception of it in the paper of two years ago by Provost Charles Phelps
of Rochester that seems to stand in an ancestral relation to the Caltech
Proposal, which apparently represents (somehow) the thinking of the
provosts of the research universities in general. As in the later
Caltech proposal, Phelps describes the LANL system solely in terms of
its librarial functions, saying nothing of its publication function, and
characterizes it, moreover, as a "forum", which is an inaccurate
description of it. (Also, I note that in the BMJ/Stanford plan for an
archive the LANL system also seems to be implicitly regarded as a forum,
since that term is used in their description of their archive as modeled
on the one at Los Alamos.) Now, the Lingua Franca characterization
above does not refer to it as a "forum", but like this misunderstanding
of what it does, it also obscures the primary publication function which
the preprint server at Los Alamos serves, and in doing so diminishes the
significance of it in a similar way.

What justification is there for regarding the LANL preprint server as a
system of publication? In a loose sense the word "publication" can be
construed simply as meaning "to make public", and in that sense anything
established as a public place that contains messages made publicly
available through it is a medium of publication. But that sense is too
loose to be useful for our purposes. We need to ground the idea of
publication in practices that are specific to disciplined intellectual
life, be it scientific or scholarly, and although we might in time
improve upon it I suggest that we can do no better at this time than to
start from a description by Joshua Lederberg, in his paper "Options for
the Future," of what he calls "the primary scientific literature",
meaning that kind of communication which the scientist as such is
rightly regarded as obligated to try to make a contribution to. He
describes it as: "the original reporting of scientific data and theory,
formulation and assertion of claims with respect to priority and the
like," remarking that "a very different set of rules applies to that
literature than to the dissemination of monographs, textbooks, novels,
biographies and so on."

After pointing out what has become a commonplace here in the AMSCI
Forum, namely, that the authors, by and large, "are totally uninterested
in royalties, which indeed have generally not been available to them.
Their gain from publication is recognition by their colleagues and the
dissemination of knowledge in the spirit of science", Lederberg makes
some further comments worth quoting at length:

> Primary literature constitutes, as it stands on the shelves
> of libraries and in other formats, an unerasable public archive.
> I think it is of the utmost importance that we not take fluidity
> too lightly, that there be a point of commitment when the
> author says: "These were my words; this is what I said we have
> done in the laboratory; these are the claims that I am making
> and they should not be tampered with at any further stage".
> Obviously they may be modified, there can be links to corrections
> addenda, further distinctions, but it should always be possible to
> reconstruct what the author is to be held accountable for.
>
> The registration of claims is a very important point - what drives
> science is the possibility of making a novel discovery. It's not a novel
> discovery if somebody else made it first, and we get very scant
> credit for even being hard on the heels of others, who had managed
> to get there a few minutes before. There are many implications of
> the allocation of scores for achieving success in that regard, but it's
> built into the structure of science as innovation that there be a
> system of registration of claims for what is new, what is different,
> what is distinctive, what was "my contribution" to the growing corpus
> of scientific knowledge and understanding.
>
> A scientific publication is a grave act to be undertaken with the
> utmost seriousness; it's an inscription under oath. To lie in a
> publication is de facto perjury, and, when that is discovered, there
> are consequences no less serious than perjury in court and so it
> should be. We shouldn't have to worry about whether an assertion
> of data from an experiment or other claims was made with other
> than the utmost seriousness on the part of the author. Otherwise,
> we would be eternally confused whether the matters presented
> deserve our attention. And the literature is a historic repository
> where the record of our scientific culture can be refreshed and
> re-examined for the purposes of history, for the purposes of
> rediscovering ancient matters, whose significance was not fully
> understood before, the establishment of links between different
> disciplines, and so forth.

http://www.dlib.org/dlib/may96/05lederberg.html

What I call "primary publication" is the act of contributing something
to the primary scientific literature, as Lederberg defines that. Given
this understanding we can differentiate the function of the LANL
preprint server system from other kinds of making public that might be
confused with it by noting that it functions as an instrument of primary
publication in just the same way that the refereed journals in that
discipline are instruments of primary publication. It has been
understood by Ginsparg himself to be that, and his perception of its
publication function seems to be corroborated by the way it is actually
used, at least in some fields.

This is what has given it its special interest as an innovation in
communication in the sciences. In each of the four papers he has
written about the preprint server system he developed at Los Alamos,
Paul Ginsparg is at pains to distinguish it from a bulletin board system
(like the usenet "news" system), presumably because that was the common
confusion that prevailed during the period of composition of those
papers (1994-1996), and I am suggesting that we need to insist upon a
similar distinction between the LANL system and a forum as well, as that
seems to be the kind of confusion which is now more common. Clarity on
these differences is essential for extending the principles of the LANL
preprint server across the disciplines insofar as that is possible.

There are at least two different reasons why the LANL preprint system
has been recognized to be of special interest. The first is that it
eliminated the rationale for the continuing existence of the "invisible
colleges" of privileged access to leading edge research in the fields
which it directly serves. This is implicit in the characterization of
it in the Lingua Franca article, and I have never seen a description of
its function which did not touch on this, explicitly or implicitly. It
is surely an important aspect of it: much of the idealism associated
with the LANL server has been based on the belief that a central ideal
of science has been actualized through this system, and it would be
foolish to brush this idealism aside as unimportant at a time when the
image of the sciences within academia is at an all-time low as regards
collegial respect, and is likely to move lower still as the commercial
buy-out of the universities and its researchers continues, as it seems
destined to do for some time to come. The LANL achievement in this
respect has been to demonstrate by its continuing existence that ideals
of pure research are still a reality in science and not a mere myth used
as a mask by people motivated only by defense of privilege and the
rewards of power, prestige, and wealth. Those interested in promoting a
more adequate understanding of science in this respect will discover in
it something of great value for their purposes.

The second basis of its significance, which is subtly masked by the
Lingua Franca description but is clearly recognized in Ginsparg's own
papers on this, is that it raises important questions -- by no means
adequately addressed and answered thus far, in my opinion -- about the
actual and legitimate function of peer review as a control factor in
scientific inquiry. If the users of the LANL system are at the leading
edge in their fields, such that the course of inquiry in the field of
physics it serves is actually being determined more by their
communication via an automated preprint system than by the refereed
journals, then what, exactly, is the peer review process for? There are
answers to that, of course, and I am by no means sceptical myself about
the importance of peer review; but at the moment I am merely pointing
out the question which the success of the LANL system has raised, which
goes right to the heart of academic faculty control systems in general,
which depend upon peer review functions in several different ways.

Back now to the confusion mentioned, which stems from the fact that in
all three cases -- the preprint server, the "usenet news" bulletin
boards, and the open discussion forums -- there is an automated
communicational process by which somebody makes something public to
others with similar interests without prior critical screening. Now,
these can be distinguished from one another by differences in the
mechanism by which what is deposited is made available: the LANL server
sends a notice that a certain message (i.e. article) is available for
retrieval; the "news" systems make messages available collectively for
reading on-line for a certain period of time, sorted by interest group
names; the list-based discussion forum distributes the messages to all
subscribers. But the important difference between the LANL system and
the other two, for the purpose of understanding the problems involved in
the extensibility of its principles, is not captured by this way of
differentiating them nor by any technical systems description. The
difference is one based on professional understanding of what is
required and appropriate in communication of this type or that and the
recognition of this in the particular case as that is shown in how
things that are said are actually treated and responded to by others.
The difference is, in short, in professional practices.

I've never made use of the usenet news boards myself because the
communicational form does not seem to me close enough to the proper
communicational form for philosophy to make it worthwhile to do so --
and other philosophers seem to feel the same -- so I won't speak
especially to that particular mode of network communication; but then I
suppose no one would now tend to confuse the LANL preprint system with a
bulletin board system, in any case. I have, however, managed a
list-based open discussion forum for some six years and have
participated in a variety of such forums with the conscious aim of
finding out what is and is not possible in such media, and one thing I
am certain of is that although open (unmoderated) forums can be of great
intellectual value, certain conditions being met, the one thing they
cannot be and should not be expected to emulate is a place of primary
publication. In fact, the mistaken attempt to use an open forum in that
way typically has the immediate effect of ruining it as a forum.
(Moderated forums such as the present one are very different in
rationale and function than those under discussion here, but there is no
need to discuss this since no one is likely to confuse this sort of
forum with an instrument of primary publication).

Lederberg's description of the special seriousness of primary
publication above, and of the commitments and responsibilities it
involves, indicates why these two very different communicational
activities should not be confused, and also suggests why it is
undesirable to encourage or tolerate ambiguity in this respect, namely,
because a preprint server and a forum server are not merely different in
purpose but are at odds with one another as regards what they are trying
to accomplish, so that to allow confusion about the function of a given
implementation of either of them is to insure failure of the
implementation to amount to anything worth while in either respect.

Characterized briefly: the purpose of primary publication is to affect
directly the ongoing process of inquiry -- hopefully for the better --
by affecting what is premised or presupposed in future inquiry on the
part of the presumptive peers to whom one's communication is addressed,
whereas the play of discussion in an open forum is at its best when one
is free to explore ideas and adopt them tentatively and trace out their
implications, criticize them, and simply abandon them at times in a much
more lighthearted spirit than is appropriate in primary publication
because it is understood that, in the communicational space of a forum,
a certain detachment from the sort of commitment and responsibility that
is entailed in primary publication is in force.

As a list manager of a lively and, I believe, successful such forum for
some six years now -- I have had much testimony to that effect and none
to the contrary, and it contains as members a substantial number of the
"establishment" figures in my special area of competence -- I have come
to see the way in which this operates in practice quite clearly, and
could spell the communicational presuppositions out in detail if
necessary. I won't get into that here, but I should remark that it is a
common occurrence for a newcomer to a forum of that sort to try to
understand it first of all in terms of primary publication practices,
even when that has been pointed out to be impertinent, and some find
they are so habituated by academic practice as to be unable to
communicate through the written word in any other mode and drop out in
frustration and shame after posting a few messages that are perfectly
acceptable there but which they can themselves only regard as badly
flawed professionally because of the rigidity of their critical
framework, which is the one appropriate to primary publication practices
rather than forum practices.

A case might be made that well-constructed and managed open forums for
intellectual life, scientific or scholarly, are quite as important as
are media of primary publication -- I personally believe that the
paucity of open forums in academia in particular has been and, even now,
continues to be a blight on intellectual life across the whole range of
academic disciplines, stunting the growth of critical control in many
fields -- but the conditions for the success of such forums are at odds
with the conditions required for primary publication: you can't do both
at once. The distinctive significance of the LANL system consists in
the fact that it is an autonomous publication system within the
sciences, functioning as a medium of primary publication just as the
peer reviewed and editorially controlled journals do. There may be
forums on the LANL website -- I've been unable to detect any forums
proper there myself -- but as regards the function of the LANL system in
respect to those fields that it originally served, at least, it is a
place of primary publication and not a forum.

I omit much I would like to say in this connection and will only make
the concluding point that if clarity is not gotten on this, the present
movement is likely to be sputtering out prematurely because of the
contradiction involved in basing it on the LANL preprint server as the
starting point and then abandoning that as an essential element in the
movement because it proves to be too problematic to incorporate in the
system being developed. The fact is there is no danger whatever that
the sciences in general and the other disciplines as well are going to
become demoralized by addiction to unregulated preprint servers
functioning as instruments of deranged primary publication. The real
problem is going to be to find fields in academia mature enough in their
professional practices to be capable of making use of such a system and
adjusting their formal peer review practices accordingly, as is
apparently being done in physics in some areas in an admirably realistic
way. Where that is impossible, open discussion forums -- not
moderated: there are better means of control than that if inquiry is
wanted -- should be encouraged and their development systematically
cultivated, with the aim of finding out, among other things, what the
situation is in the given field as regards publication practices and
communication generally that puts that out of the question. Do not
expect such forums to transmue into media of primary publication,
though.

As a model the LANL preprint server is a vanguard, which is not the same
as a paradigm to be emulated with no understanding of what the
conditions for successful emulation actually are. For that
understanding I suggest, first of all, a very close reading of Paul
Ginsparg's papers (available at the xxx website), with special attention
to the detail of his development of it as a primary publication system,
with the question in mind of whether this or that other field is really
mature enough yet in its normal communicational practices to take
advantage of what the Ginsparg system offers. There could be many such
fields or few: I have no opinion on that. The chief point of the
present message is that it is important, in any case, to avoid
explaining it away as something else -- such as a discussion forum or a
bulletin board, or, as in the Lingua Franca piece, as a sort of
preparatory zone relative to the goal of getting one's papers published
in impeccable form. That is not what the LANL server is all about, and
its principles can never be extended effectively to other disciplines
with that sort of misunderstanding of it.

--
Joseph Ransdell  <ransdell_at_door.net> or <bnjmr_at_ttu.edu>
Dept of Philosophy   Texas Tech Univ.  Lubbock TX 79409
(806)  742-3158 office    797-2592 home    742-0730 fax
ARISBE:Peirce Telecommunity  http://www.door.net/arisbe
   http://www.door.net/arisbe/homepage/ransdell.htm
Received on Wed Feb 10 1999 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:45:34 GMT