Re: The True Cost of the Essentials (Implementing Peer Review)

From: Stevan Harnad <>
Date: Sat, 15 Dec 2001 23:56:01 +0000

On Sat, 15 Dec 2001, Bernard Lang wrote:

> We can publish first, and review or copy edit later, in whatever
> order is convenient, or never if no one wishes to do it.

We can publicly archive first (let's reserve the term "publish" for
something more than this mere vanity-press, lest it lose its meaning)
and then we can submit that unrefereed preprint to an established
journal for peer review. (Why established? Because otherwise you have
no way to know what quality-standards have been met by their having
accepted it for publication!)

Or, we can leave the paper forever as merely a publicly archived,
unrefereed preprint.

The primary objective of this Forum, however, is to attain free online
access to the entire full-text contents of the peer-reviewed corpus of
20,000 refereed journals. Vanity self-archiving of unrefereed preprints
does not meet that objective. Online access to unrefereed preprints is
merely a bonus, an extra, not an alternative way of meeting the objective
of attaining free online access to the peer-reviewed corpus.

"Self-Archiving Refereed Research vs. Self-Publishing Unrefereed Research"

> I do not care if, when, and how reviewing has been done... all I need
> to know is whether it has been done, and by whom or what group, and
> maybe even have the comments.

(This is a bit confusing, as "if" is synonymous with "whether," mais

Whether the "it" has been done, and "by whom," for our purposes, is the
question of which known, established quality-controller and certifier
(i.e., which journal) has peer-reviewed and accepted the paper. That
tags its level in the quality hierarchy, and those tags are critical
for navigating the enormous literature for busy researchers who would
rather not spend their time reading or trying to build upon material of
uncertain quality. This kind of reliable filtering cannot be done on
an ad hoc basis (any more than eggs can be certified on an ad hoc
basis: the egg-graders have to establish their reputations).

And comments are always welcome, but they are a luxury. See:

> With that I am a big enough boy to make my own decisions. Choosing
> a journal is just choosing a set of reviewers. Why should I do it
> before I know what papers I'll be looking for. Why not consider a
> bunch of papers and then decide which types of reviews I'll consider
> adequate (for example depending on how selective I need be).

Because there are only so many hours in the day, and an awful lot of
stuff is written. I would rather have trusted quality filters in
advance, not after I have committed my time! and I'd rather have a
literature already written with the foreknowledge (on the part of its
authors) that it will have to answer to peer review. And for the peer
reviewers to be able to certify that I can trust a paper, I first have
to know I can trust the peer review. So its quality level must have
been reliably demonstrated in advance.

In other words, I need journals.

> And why should papers have only one type of reviewing, when they
> are so many different publics with different needs, even within the
> not for profit litterature.

Because peer-review is a scarce, over-farmed resource; because peers
review for free; because one review is more than enough for most
papers; and because pre-certification peer review is not the same a
post-certification peer commentary...

"A Note of Caution About 'Reforming the System'"

    Harnad, S. (1997) Learned Inquiry and the Net: The Role of Peer
    Review, Peer Commentary and Copyright. Learned Publishing 11(4)
    283-292. Short version appeared in 1997 in Antiquity 71: 1042-1048.
    Excerpts also appeared in the University of Toronto Bulletin: 51(6)
    P. 12.

    Harnad, S. (1998) The invisible hand of peer review. Nature
    [online] (c. 5 Nov. 1998)
    Longer version:
    Harnad, S. (2000) The Invisible Hand of Peer Review, Exploit
    Interactive, issue 5, April 2000>:
Received on Sat Dec 15 2001 - 23:56:11 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:20 GMT