Searching and Thinking About Searching JSTOR

Author(s): Lisa Gitelman

Source: Representations , Vol. 127, No. 1 (Summer 2014), pp. 73-82
Published by: University of California Press
Stable URL: https://www.jstor.org/stable/10.1525/rep.2014.127.1.73

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

University of California Press is collaborating with JSTOR to digitize, preserve and extend access
to Representations

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
LISA GITELMAN

Searching and
Thinking About Searching JSTOR

I N M A R C H 2 0 1 3 T H E O N L I N E repository JSTOR unveiled its

new Beta Search. Perhaps that is how you, reader, have landed on this page
and now read these words, since the journal Representations is part of JSTOR.
According to its ‘‘about’’ pages, Beta Search is aimed at helping ‘‘you find
what you need more quickly’’ than before. And JSTOR wants to know if it is
working: ‘‘Are you seeing better results?’’ @JSTOR tweets to interlocutors,
while behind the scenes JSTOR’s architects plan to improve its new interface
and search algorithms, ‘‘informed by usage analytics and feedback from
those who try’’ the new search function. Improvements in the search inter-
face have made it easier to narrow or broaden searches in different ways,
while improvements in the results interface have made it easier to tell what it
is your searching has turned up. The underlying search algorithms have
been improved partly on the basis of prior search activity, so that all of the
JSTOR searching you’ve done in the past is today helping to make JSTOR
‘‘more forgiving’’ of things like spelling and punctuation errors; it can now
‘‘accurately interpret’’—that is, predict—more of what its users are search-
ing for. Meanwhile new topic-modeling technology is getting users results
that are less structured by the ways JSTOR sorts journals by discipline and
disciplines by journal and more structured by computational analysis of the
vast JSTOR corpus.1
JSTOR—the name is an acronym for ‘‘journal storage’’—is an indepen-
dent nonprofit entity that was dreamed up in 1993 as a Mellon Foundation
project by Mellon’s then president, William G. Bowen, in the same year that
Adobe Systems developed the first iteration of PDF (the portable document
format) and the same year that the National Center for Supercomputing
Applications released its Mosaic browser, whence Netscape.2 In case you’ve
missed it since it first went live online in 1997, JSTOR makes scholarly journal

a b s t r a c t Digital resources are helping to change the ways scholars and students work, but
they must also be helping to shape the work that gets done. Taking JSTOR as an example, we might
ask about the discursive power of the database. How is using an online resource for research acceding
to unnoticed assumptions that underlie the construction of that resource? R eprese ntation s 127.
Summer 2014 © The Regents of the University of California. ISSN 0734-6018, electronic ISSN 1533-
855X, pages 73–82. All rights reserved. Direct requests for permission to photocopy or reproduce
article content to the University of California Press at http://www.ucpressjournals.com/reprintinfo.
asp. DOI: 10.1525/rep.2014.127.6.73. 73

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
 literature available to its subscribers, mostly colleges and universities, and has
become—especially for scholars with access—both welcome and necessary as
a research resource, particularly within the humanities and social sciences.
Like many online resources of this sort, JSTOR is fully searchable because it
contains page images linked to an invisible ‘‘layer of [admittedly] ‘dirty’ full-
text.’’3 Users search an unseen trove of ASCII (the American Standard Code
for Information Interchange character encoding scheme), structured by
a trove of metadata, and then a list of search results points them toward preview
images of full texts, where users can select to view digital or digitized docu-
ments as PDF page images that can be downloaded. Full-text search and (espe-
cially) stable page image results help to signal the humanistic orientation of
this research resource, since they cater to users who build arguments partly
from readings that use direct quotations and so need to cite by page number.
In light of its extraordinary success—its invisible centrality to so much
scholarly inquiry—JSTOR begs of databases many of the same questions that
Michel Foucault once directed toward the concept of the archive. According
to Foucault, an archive is less a collection of texts than a system governing
statements.4 Texts possess meanings, yes, but systems produce and delimit
knowledge. How does a database like JSTOR work as (or as part of) a system?
What, in short, is JSTOR’s discursive power? How is searching JSTOR like
using a stacked deck, now being restacked recursively as searching is ana-
lyzed to predict further searching? For many of us JSTOR has put an
unimaginably vast corpus of journal articles at our fingertips, all but zeroing
our trips to the library for journal literature while at the same time opening
our eyes to a corpus both wider and deeper than snail methods once
allowed. But how has using JSTOR helped to change not just the methods
but also the very substance of scholarship itself? How do new tools for
knowledge production leave their marks on the knowledge they help us
produce? Difficult questions all, made more so by the fact that JSTOR pre-
sents a moving target, as it continues to add journals and now books, as well
as to add issues to the journals it collects, growing and improving its own
functionality at the same time that it benefits from ongoing improvements
to search and web technology. To date it contains the back files of more
than two thousand academic journals.5 JSTOR keeps getting better and
better. Reader, make no mistake: I ♥ JSTOR.
Originally conceived as a way to rescue university libraries from their
space crunch and the debilitating costs of expansion, JSTOR was born partly
of chronic bibliometric pressure and partly of the economics of the long tail.
As both their number and their respective back files increase, journals are
more and more expensive to store at the same time that older issues—
essential to increasingly specialized scholarly inquiry—are consulted less
and less. Microform was for many years promoted as a panacea wherever

74 Representations

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
storage costs were concerned, but by 1993 digitization seemed to offer
a storage medium that was more compact and less cumbersome, its contents
both searchable and transmissive across networks, whether those networks
were the local CWIS (campus-wide information systems) of the day or the
expanding Internet beyond.6 Adobe’s initial promotion of PDF technology
captures the sentiment in Beyond Paper, a booklet it published, ironically
enough, on paper. The opening pages of Beyond Paper include an isolated
panel of italicized text, suggestive but unexplained: ‘‘A major university library
is spending $1.5 million on a supercomputer to electronically store the contents of more
than 40,000 deteriorating old books. The cost of building a library addition to house
those books would be $20 million.’’7 The implied accessory to Adobe’s distinc-
tion between contents and books is the apparent destruction of the latter.
JSTOR has been much more careful to respect its sources. It has defined
archiving as part of its mission and business model, noting that it works with
the California Digital Library and Harvard Depository to preserve ‘‘multiple
copies of the original print publications underlying the [digital] archives so
that they are available for re-digitization as well as other unanticipated
needs.’’ Preservation of the digital content, meanwhile, has always been
done in house, making JSTOR an early example of third-party archiving.
JSTOR not only preserves image scans in the TIFF G4 (tagged image file
format, group 4 compression) archival standard, it has also long agreed to
provide subscribers with full copies of the database in the unlikely event of
its own demise.8
JSTOR’s planners and architects decided on a database that offered
‘‘perfect fidelity to the original pagination and layout, as well as an accurate
reproduction of the text.’’ They did this for two reasons: so that scholars
could quote and cite with confidence just as they always had, and because
they worried that the alternative (text, raw or marked up, like the Lexis-
Nexis database) might violate the copyrights of individual scholar-authors.9
It was already firmly established that publishers—rather than their scattered
authors—could give permission to microfilm whole journal runs, so JSTOR
wanted a ‘‘microfilm-like rendition’’ with digital capabilities.10 JSTOR began
digitizing from library copies, but quickly switched to publishers’ warehouse
copies where they existed, supplementing these when necessary, assembling
each digital text as an ideal work, the likeness of a continuous run of each
journal. In the language of textual scholarship, JSTOR offers noncritical
facsimile editions of the journals it contains, basing each upon an inexplicit
assemblage of extant tokens according to the varied accessibility of fair
copies. JSTOR journals are ‘‘noncritical’’ editions because they don’t seek
to establish—as it is called—the text through bibliographical enumeration
and description.11 Instead JSTOR perforce establishes a text for every jour-
nal it includes by digitizing one copy (though today plenty of content comes

Searching and Thinking About Searching JSTOR 75

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
 to JSTOR ‘‘born digital.’’) Where once scholars around the world used and
cited unique, local copies of printed journals, today scholars everywhere use
and cite ‘‘the same’’ copy as a remote file online. To the extent that the
concept makes any sense in the digital realm, sameness now inheres in the
production, storage, and ownership of data on a server (and its mirror
servers) and not in the locality of its reception on a client.12
If JSTOR has thus been part of a profound reshaping of what might be
called geo-bibliography, it has nonetheless acceded to the existing copyright
regime. Whatever its other effects and affordances, in other words, JSTOR
works as (or as part of) a commercial system. Its contents are structured by
ownership in at least three respects. First, again, its use of stable page images
is in some measure a response to copyright concerns. (By contrast, its some-
time competitor Project Muse—started by copyright-holder Johns Hopkins
University Press—offers both HTML and PDF versions of journal articles.)
Second, much to the dismay of advocates for open access, the database is
proprietary, and users must subscribe or be affiliated with a subscribing
institution.13 JSTOR generally charges its subscribers based on types and
sizes of institution, with different rates for different countries around the
world.14 And third, JSTOR continues to update the journals it collects
according to its ‘‘moving wall’’ agreements with publishers. Current issues
(anywhere from one to five years’ worth) remain proprietary and apart from
JSTOR—unless included in its recent ‘‘Current Scholarship Program’’—but
publishers send them along for inclusion in the database once they lose
their currency.15 JSTOR thus protects the interest that rights holders—the
publishers—have in new issues, while its growing collection of older issues
enhances the value of the journals and JSTOR mutually. Implicit calendars
of commercial obsolescence and archival incorporation now supplement
the traditional, explicit calendar of periodical publication, offering yet
another example of the ways in which, as Will Straw puts it, ‘‘new technol-
ogies have consistently rendered the past more richly variegated and
dense.’’16 JSTOR’s original mission to save space in the stacks has been all
but forgotten by users and subscribers alike, but the metaphor of the mov-
ing wall helps to keep the imagination of additional shelf space part of the
logic of JSTOR.
JSTOR’s not-for-profit character, like its ‘‘.org’’ domain name, works as
a partial abnegation of its commercial character, and in this JSTOR is
a descendant of a long tradition of gentlemanly endeavor. Like the ‘‘Gentle-
men’s Agreement’’ of the 1930s, that is (an informal agreement between
scholars and publishers about fair use copying for the purposes of research),
JSTOR works forever in vain to tailor the resemblance of a gift economy to fit
the market economy in which it operates.17 Nowhere are the pitfalls and
discomforts incumbent in this project more evident than in the case of Aaron

76 Representations

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
Swartz. In the fall of 2010 hacktivist Aaron Swartz used an IP (Internet pro-
tocol) address at MIT to download about 80 percent of JSTOR, apparently as
part of an effort to set its contents free. JSTOR’s systems administrators
watched with alarm as automated ‘‘pdf scraping’’ by ‘‘the bad guys’’ or an
‘‘offending scraper’’ pulled down hundreds of thousands of journal articles.18
When Swartz was ultimately apprehended and identified as the culprit,
JSTOR agreed not to pursue civil charges against him if he would return its
data without copying or releasing it. The US Attorney’s Office was not as
forgiving, and Swartz committed suicide with the criminal case against him
still pending. Suddenly in the spotlight, JSTOR sought to explain itself and its
mission to the public. In the face of ‘‘significant misuse of our database,’’
JSTOR said, it did not have any interest in prosecution, only ‘‘in securing
[our] content.’’ Pointing to the importance of its idealistic mission ‘‘to
expand access to scholarly knowledge worldwide and to preserve it for future
generations,’’ JSTOR has explained that it can’t fulfill this mission without
maintaining its relationships with publishers.19 At the same time, JSTOR has
been quick to add, it continues to expand access—including free access—in
a variety of ways. Early journal content (published pre-1923 in the United
States or pre-1870 elsewhere) is now free, for example, and all of JSTOR is
freely accessible to its subscribers in Africa. The upshot: though responsive to
its critics in some measure, JSTOR has a business model, and it’s sticking to it.
There is a book about JSTOR that offers a thoroughgoing account of the
origins of this business model. The book is called, naturally enough, JSTOR:
A History (Princeton, 2003). ‘‘Why a book on JSTOR?’’ the introduction asks.
A short answer is that the Mellon Foundation wanted one. Author Roger C.
Schonfeld was a Mellon employee who previously worked for JSTOR and
now works for the related ITHAKA SþR academic research and consulting
service. The book has been widely and generously reviewed, though both of
the flattering blurbs on the dust jacket are by individuals mentioned in
Schonfeld’s acknowledgements; ‘‘Each read the manuscript, which
benefited greatly from their suggestions’’ (xxi).20 I mention the self-
endorsing qualities of JSTOR the book not as a criticism but rather because
JSTOR the database can have the same feel. One early citation analysis—of
JSTOR by JSTOR—indicated, ‘‘Of the fifteen disciplines making up Arts &
Sciences I [the first collection of journals completed], in two of them JSTOR
titles constituted about 75 percent of the citations; in another four, over 40
percent of the citations’’ (207). This means that articles that have ended up
in JSTOR cite other articles that have ended up in JSTOR, suggesting either
the importance of its collections or the published coherence of the schol-
arship that JSTOR represents, or both. Today 1.8 million of JSTOR’s 8.3
million documents ‘‘cite or are cited by other documents in the JSTOR
corpus.’’21 Now that inter-JSTOR citations are hyperlinks, users can click

Searching and Thinking About Searching JSTOR 77

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
 around the literature, gaining a sense of connective plenitude that may lead
some users to a false sense of completeness. Like so much online—like
Google—an overwhelming amount of data can start to seem like everything
there is.
If students or scholars stop with JSTOR—without seeking additional
sources or pondering its contents and their limits—then JSTOR makes
them the victims of its success. It becomes the system governing statements.
Using the new Beta Search interface, you can now search ‘‘all content’’ or
‘‘only content I can access.’’ Each option omits the implied phrase ‘‘in
JSTOR,’’ while the choice between the two will, JSTOR hopes, prompt you
to an awareness of the commercial system in play. Whichever option you
select, there are features of JSTOR that hint otherwise at its discursive
power. For one thing, the database represents another obvious instance
of both the Anglophone web and the worldwide hegemony of the US acad-
emy. The ‘‘About JSTOR’’ pages at jstor.org once appeared in Chinese,
German, Italian, Japanese, Russian, Spanish, French, and Korean (today it’s
just English), but the JSTOR collection itself has always been primarily
English-language and largely—though not entirely—North American. Yet
by 2009 JSTOR had more participating institutions outside the United
States (3,070) than within it (2,643).22 Today its website lists 10,604 sub-
scribing institutions in 150 nations around the world.23
If its geopolitical tilt seems obvious, JSTOR has other biases that can be
harder to pick out. Most notably, the development of JSTOR has entailed
a certain model of disciplines and of disciplinarity, first by dint of the way it
initially pursued ‘‘core’’ journals to form a ‘‘critical mass,’’ and second by
dint of its ‘‘curated collections,’’ which bundle journals for different sub-
scribers. To begin, JSTOR decided to collect journals by discipline, and it
started with disciplines for which it would be easy to identify what it saw as
‘‘the most important titles’’ (JSTOR: A History, 19). That is, JSTOR began
with the hopeful assumption that disciplines—at least as they are reflected
by journals—are at some level plausibly discrete. First up were the disci-
plines of economics and history. (Bowen, the genius behind JSTOR, is an
economic historian.) These JSTOR considered ‘‘stable’’ disciplines in 1994
by comparison to something like literature, which was deemed ‘‘too turbu-
lent for a simple selection of 5–10 [pilot] journals,’’ since, as one consultant
put it, ‘‘the boundaries of, and the terrain within’’ literary studies remain ‘‘in
flux’’ (20). Messy disciplines, or disciplines that challenge the discrete model
of what a discipline is, would have to wait a little while. Interdisciplinarity was
a desideratum engaged specifically on disciplinary terms at first, as ‘‘fields that
directly foste[r] interdisciplinary work, most notably statistics, were consid-
ered to be especially important’’ (201). The first collection of journals to be
made public, ‘‘Arts & Sciences I,’’ included fifteen disciplines—among them

78 Representations

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
history, economics, finance, literature, and statistics—and 117 titles. Only 7 of
these first 117 titles—approximately 6 percent—could be counted under
more than one discipline (200).
Intent on library storage problems, JSTOR opted to start digitizing ‘‘the
most central, important, widely used’’ or ‘‘core’’ journals of a discipline first,
the so-called flagships and clusters around them to form a ‘‘critical mass,’’
rather than digitizing ‘‘more specialized and less commonly collected’’ jour-
nals (17, 99). That way more libraries would achieve more savings, even if
fewer total researchers might gain access to more journals. As sensible as this
decision seems, it ran counter to much of the prevailing wisdom on elec-
tronic resources during the extended moment when the World Wide Web
was becoming the dominant platform for their distribution. Scholars and
scholarly institutions were bent on publishing rare resources, not the most
common ones, as they sought to leverage the celebrated ‘‘democratizing’’
potential of the web against the backdrop of dot.com hype and start-up
speculation. (Web editions that collected far-flung or unique resources were
typical, making individual William Blake prints or Thomas Edison holo-
graphs available to students and scholars everywhere, for instance, mitigat-
ing the costs of travel among archives.) Rare resources are used by small
numbers of specialists, while large numbers within a field use its common
journals. Here again is an implicit model of disciplines as core/periphery,
a model that emphasizes centers of gravity, if you like, over what Bruno
Latour calls ‘‘centers of calculation.’’24 With its attention on back files, that
is, JSTOR was of necessity attending to the inertial coherence of disciplines
rather than to the scattered loci of disciplinary change, the cutting edges
and moving parts of knowledge production, circulation, and exchange.
Current issues weren’t included, but neither were new journals, small-
circulation journals, irregular journals, or electronic journals. (The hard
sciences tended to be excluded for other reasons, as did journals published
by Johns Hopkins, which at the time appeared only within the competing
Project Muse.)
Initial incarnations of JSTOR thus helped at some level to conserve and
reify the traditional disciplines, acceding and thereby adding to the inertial
weight they accrue by dint of institutional contexts—publishing companies,
learned societies, departments—and the manifold ways in which scholars,
administrators, and librarians internalize and perpetuate them. We must
speculate, however, that JSTOR’s model of the disciplines and of disciplin-
arity is of significantly less moment today, now that JSTOR includes thou-
sands of journals, not 117. As JSTOR has grown, its disciplines have grown
less discrete. Its model, even if still in place, has less discursive power. More
journals simply mean more overlapping domains of inquiry, particularly
as JSTOR has returned to disciplines and added to their range. That said,

Searching and Thinking About Searching JSTOR 79

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
 JSTOR continues to serve as a curator of journal collections, bundling them
into ‘‘Arts & Sciences, Parts I to VII,’’ as well as other, discipline-specific
collections, like ‘‘Business I,’’ ‘‘Business II,’’ and ‘‘Ecology’’ and ‘‘Botany.’’
To the extent that some subscribers opt for selected collections—whether to
save money or in light of a specific institutional mission—an older, smaller,
and stodgier JSTOR remains very much in play, its model of disciplinarity
locally and invisibly in force at those institutions. Speculative, yes, but the
situation is potentially reminiscent of one described by William St. Clair in
The Reading Nation in the Romantic Period.25 St. Clair argues that licensing laws
—ownership—encouraged the continued, cheap republication of folk
knowledge for the British underclass at the same time that a small, elite audi-
ence had access to change, to modernity, in the form of newer publications
like romantic literature. The economics of access may induce a sort of cultural
lag or what St. Clair calls ‘‘lock-in’’ for those with the least resources, a hazard
JSTOR clearly aims to mitigate with its different subscriber rates and its
special initiatives for reduced-rate subscriptions in the developing world.
There is no denying what a stunning accomplishment JSTOR is and
what an opportunity it provides for further thought both about the ways
knowledge is produced within and across disciplines and about how disci-
plines change. Computational analysis will be particularly welcome, and
JSTOR is encouraging it with an initiative called ‘‘Data for Research,’’ which
allows researchers to download specific datasets for analysis. Even apart
from that analysis, however, using JSTOR—no matter how adept at search-
ing you are or are not—must remain an encounter haunted by unknowns.
What are the best ways to glimpse the history and logic of a database and to
reckon its discursive power? How is searching helping to change what we
know and how we know, in this our ‘‘epoch of potential memory,’’ when
knowledge is born somehow of information, that curious and ‘‘morselized’’
substance we fish from databases?26

Notes

1. Tweets from JSTOR to users, e.g., @JSTOR to Lisa Dotzauer@AtomicDucks, 23

May 2013, 9:35 a.m., https://twitter.com/search?f¼realtime&q¼%40atomic
ducks&src¼typd; @JSTOR to Mari@MariPari04, 24 June 2013, 7:41 a.m.,
https://twitter.com/search?q¼%40maripari04&src¼typd.
2. See Roger C. Schonfeld, JSTOR: A History (Princeton, 2003). Pages from Schon-
feld will be cited parenthetically in the text after his book is described.
3. Ibid., 28–29.
4. Michel Foucault, The Archaeology of Knowledge (1972; reprint, New York, 1982),
part 3.

80 Representations

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
 5. JSTOR, ‘‘About,’’ http://about.jstor.org/10things.
6. See Eugene B. Power, Edition of One: The Autobiography of Eugene B. Power, Founder
of University Microfilms (Ann Arbor, 1990), 154. As Schonfeld, JSTOR: A History,
has it, ‘‘Making use of technology to save space and reduce costs offered a strik-
ing combination of Bowen’s background, librarians’ needs, Mellon’s mission,
and the blossoming technology of the time’’ (16).
7. Patrick Ames, Beyond Paper: The Official Guide to Adobe Acrobat (Mountain View,
CA, 1993), 8.
8. Quoted is JSTOR, ‘‘Preservation,’’ http://about.jstor.org/content/preserving-
scholarship. On archiving as mission and element of the JSTOR business
model, see Schonfeld, JSTOR: A History, 137, 179, 358–61.
9. Schonfeld, JSTOR: A History, 28, 161.
10. Ibid., 72, 39.
11. D. C. Greetham, Textual Scholarship: An Introduction (New York, 1994), 347. That
said, there were plenty of tough decisions to make and a preproduction process
that resembled that for microfilm editions. Should JSTOR digitize the ads in
journals? (Yes.) Should it digitize conference programs and directories when
journals were the organs of learned societies? (Sometimes.) See Schonfeld,
JSTOR: A History, 89–90, 156–57.
12. As Matthew G. Kirschenbaum writes, ‘‘One can, in a very literal sense, never
access the ‘same’ electronic file twice, because each and every access consti-
tutes a distinct instance of the file’’; ‘‘The.txtual Tradition,’’ in Comparative
Textual Media, ed. N. Katherine Hayles and Jessica Pressman (Minneapolis,
2013), 60.
13. The case for fully open access is made eloquently and persuasively by John
Willinsky, founder of Open Journal Systems, The Access Principle: The Case for
Open Access to Research and Scholarship (Cambridge, MA, 2009).
14. See, e.g., JSTOR’s, ‘‘Arts & Sciences I,’’ ‘‘Fees & Classification,’’ http://about.
jstor.org/content/arts-sciences-i#tab-fees. Pages like this one let potential sub-
scribers pick the ‘‘community’’ they belong to: as of this writing the communi-
ties include Canadian universities and four-year colleges, community colleges,
government and nonprofit research institutions, UK universities and four-year
colleges, US universities and four-year colleges, and universities and four-year
colleges from all other countries.
15. The small Current Scholarship Program (more than 270 journals as of this
writing) is an effort to partner with publishers to produce issues electronically;
see JSTOR, ‘‘Content on JSTOR,’’ ‘‘Current Journals,’’ http://about.jstor.org/
content-on-jstor-current-journals.
16. Will Straw, ‘‘Embedded Memories,’’ in Charles R. Acland, ed. Residual Media
(Minneapolis, 2007), 12.
17. On the Gentlemen’s Agreement see Robert C. Binkley, Manual on Methods of
Reproducing Research Materials (Ann Arbor, 1936); see also Peter B. Hirtle,
‘‘Research, Libraries, and Fair Use: The Gentlemen’s Agreement of 1935,’’
preprint, 20 March 2006, Cornell University Library, eCommons@Cornell,
http://hdl.handle.net/1813/2719.
18. JSTOR has published the emails it handed over as evidence for U.S. v. Aaron
Swartz, from which these phrases were taken, JSTOR, ‘‘United States vs. Aaron
Swartz,’’ http://docs.jstor.org/documents.html.
19. Public statement of 19 July 2011, JSTOR, ‘‘JSTOR Statement: Misuse Incident
and Criminal Case,’’ http://about.jstor.org/news/jstor-statement-misuse-incid
ent-and-criminal-case.

Searching and Thinking About Searching JSTOR 81

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms
 20. Today Amazon.com contains 2 five-star ratings. A reviewer named ‘‘rlosh4’’
once posted this review too, now gone (accessed in September 2009),
This book tells it all: JSTOR CEO Rudy ‘‘J’’ Sanchez’s ascent from an undereducated but
determined delivery man to a savage corporate sociopath who would stop at nothing to
get the academic journals he wanted, once infamously strangling an up-and-coming VP
in the back of his DeLorean. The plot thickens when Sanchez realizes that nobody reads
the articles on JSTOR and begins digitizing HUMAN ORGANS. Drugs, blackmail, lust,
illegal immigrants, cataloguers who whisper ‘‘F*ck it!’’ and place everything in reverse
alphabetical order—you name it, and the scholarly journal digitization world has seen it.
Includes the famous, formerly-banned passage that describes Sanchez sautéing a sedated
man’s brains. Read it if you can stomach it. FIVE STARS.

21. J. D. West et al., ‘‘The Role of Gender in Scholarly Authorship,’’ PLoS ONE 8, no.
7 (2013): e66212, doi:10.1371/journal.pone.0066212.
22. This is information from September 2009, JSTOR, http://www.jstor.org/page/
info/about/organization/participantLists/participantsAll.jsp, a page no longer
available.
23. See JSTOR, ‘‘JSTOR Institutions,’’ http://about.jstor.org/jstor-institutions.
24. Bruno Latour, Science in Action: How to Follow Scientists and Engineers Through
Society (Cambridge, MA, 1987), chap. 6.
25. William St. Clair, The Reading Nation in the Romantic Period (Cambridge, 2004),
77–79.
26. Geoffrey C. Bowker calls the present ‘‘the epoch of potential memory,’’ in
Memory Practices in the Sciences (Cambridge, MA, 2005), 30. On ‘‘morselized’’
information see Geoffrey Nunberg’s ‘‘Farewell to the Information Age,’’ in his
edited collection, The Future of the Book (Berkeley, 1996), 103–38.

82 Representations

This content downloaded from

49.150.2.184 on Wed, 08 May 2019 02:53:19 UTC
All use subject to https://about.jstor.org/terms