Abstract

The Belgian bibliographer and independent scholar Michel Ryckx created and maintained the world’s most extensive bibliography of Thomas Pynchon scholarship on his website Vheissu.net from 2002 through 2022. This article introduces a new resource, the Thomas Pynchon Online Bibliography (TPOB), which transforms Vheissu.net’s extensive Pynchon bibliography into an open bibliography on Zotero, a free, open-source, and widely-used reference management software, with bibliographic metadata for each item, and extensively updated with entries and additional metadata for its 1.0 release. TPOB can assist scholars in locating Pynchon studies on specific topics, by specific authors, in specific languages, etc. The TPOB dataset also supports the investigation of novel insights into Pynchon studies, and may contribute to contemporary bibliometric literary studies more broadly. This paper presents exploratory experiments on the Pynchon studies metadata in TPOB including studies by year and page count, text-to-commentary ratio, formal features of bibliographic titles, intertextual fields, and semantic web.

Keywords

bibliography, bibliographic, metadata

How to Cite

Ketzan, E., (2025) “A Bibliometric Overview of Pynchon Studies: Introducing the Thomas Pynchon Online Bibliography (TPOB)”, Orbit: A Journal of American Literature 12(1). doi: https://doi.org/10.16995/orbit.23348

This project is dedicated to the memory of Michel Ryckx

Introduction

The Belgian bibliographer and independent scholar Michel Ryckx created and maintained the world’s most extensive bibliography of Thomas Pynchon scholarship on his website Vheissu.net from 2002 through 2022. In addition, Vheissu.net contained an elaborate index of the named entities in Pynchon’s texts, painstakingly compiled by Ryckx, along with page number locations, a semantic classification system, and metadata of related terms, all according to Ryckx’s own classification. Over the years, Vheissu.net’s resources were a great aid to my own scholarship on Pynchon and the scholarship of many across the globe. Ryckx also co-authored a book chapter on “Internet Resources” in Thomas Pynchon in Context from Cambridge University Press (Ryckx and Ware, 2019) and presented at the International Pynchon Week conference. Evidence of his generous assistance to scholars can be found in author acknowledgements such as Hänggi (2020). Ryckx unfortunately passed away in 2022, but before his passing transferred the data from Vheissu.net to a number of scholars so that others might preserve and benefit from its resources in the future.

This article introduces a new resource, the Thomas Pynchon Online Bibliography (TPOB), which transforms Vheissu.net’s extensive Pynchon bibliography into an open bibliography on Zotero (a free, open-source, and widely-used reference management software),1 with bibliographic metadata for each item and extensively updated with entries and additional metadata for its 1.0 release. Through simple web-based or application-based searches, TPOB can assist scholars in locating Pynchon studies on specific topics, by specific authors, in specific languages, etc. The TPOB dataset also supports the investigation of novel insights into Pynchon studies and may contribute to contemporary bibliometric literary studies more broadly.

TPOB 1.0 is presented with the caveat that it does not claim to meet the rigorous editorial standards of e.g. Mead (1989); while hundreds of its entries have been added and/or checked for accuracy, many hundreds more rely on Vheissu.net’s static bibliographic information. Nonetheless, TPOB offers the benefits of improved bibliographic resource location and extensive, structured metadata, following the FAIR Guiding Principles for scientific data management and stewardship: Findability, Accessibility, Interoperability, and Reusability (Wilkinson et al. 2016). I present initial formal descriptions of TPOB and exploratory inquiries below, and I hope that scholars will undertake many more. I have undertaken this task in honor of Michel Ryckx and his outstanding contribution to Pynchon studies. TPOB is available on Zotero:

https://www.zotero.org/groups/5470223/thomas_pynchon_online_bibliography/library

A single file containing the contents of TPOB 1.0 will also be archived here at Orbit, as well as the author’s GitHub page.

“He spent the rest of the day upstairs among the T.W.I.T. library stacks, trying to reduce his ignorance some.” — Against the Day (2006, pg. 685).

Pynchon is one of the most studied contemporary authors, and compiling and updating any Pynchon bibliography has proven a never-ending task. Already in 1978, William Plater remarked upon “the countless articles and monographs” on Pynchon (pg. ix), and Khachig Tölölyan (1979) counted around 70 articles, 3 books, 2 anthologies, and numerous PhD dissertations. In 1989, Bernard Duyfhuizen counted an additional 250+ articles, 19 books, 5 anthologies, and 67 dissertations, and Clifford Mead’s Thomas Pynchon: A Bibliography of Primary and Secondary Materials (1989) provided the first dedicated, comprehensive Pynchon bibliography. More recently, The New Pynchon Studies (2019) and Thomas Pynchon in Context (2019) included select general bibliographies, and the online publication of the archive of the Pynchon Notes journal by Orbit (Eve et al. 2016) was a major contribution to Pynchon bibliography as well.

Vheissu.net aimed to fill the gap for a comprehensive, online, continuously updated Pynchon bibliography, and as of 2022, Vheissu.net listed 119 monographs and essay collections devoted to Pynchon, 506 books chapters and book sections, 1,532 academic journal articles and reviews, 178 articles in non-academic “cultural journals”, 445 popular press book reviews, 11 conferences, and 270 presentations. Although Vheissu.net went offline at the end of 2022, Vheissu.net’s bibliography was archived by Archive.org as HTML and PDF.2

The term bibliometrics is attributed to Alan Pritchard (1969), and bibliometric studies have multiplied across numerous disciplines, especially in natural sciences, for aims including research performance/impact evaluation (see e.g. Wallin 2005, Bredahl 2022b) and tracing subject matter trends through descriptive analysis of bibliographic indicators and network analysis (see Sugimoto and Larivière 2018). A succinct summary of bibliometrics presents a challenge, as, per Laura Bredahl, there are “almost countless bibliographic tools and technologies available today” (2022a). In the humanities and literary studies specifically, bibliometric studies are far behind the natural sciences in number but recently growing. Björn Hammarfelt summarized bibliometric research in the humanities as “slowly maturing”, and that “the field is gradually moving from analyzing coverage to a new line of inquiry that tries to understand the humanities on its own terms” (2016, pg. 115). Hammarfelt notes how “historically, bibliometric research on the humanities has focused mainly on the inadequate coverage of publications by humanities scholars in available citation data-bases,” (2016 pg. 17) and indeed, one justification for TPOB is that a large number of Pynchon studies are not indexed, or indexed with minimal metadata, by leading online bibliometric platforms.

A fuller literature review of literary bibliometrics is beyond the scope of this paper; as Tanselle writes, “the various kinds of scholarly endeavor often referred to as ‘bibliography’ are interrelated, and the history of any one of them necessarily impinges on the history of the others (1988, 34). A notable study to mention is “Research Trends in Postmodernism: A Bibliometric Analysis” by Hariharasudan et al. (2022), which presents relevant data on the chronology and source country of academic publications on “postmodernism”, with the major limitation of the conflicting definitions of “postmodernism”, which the authors acknowledge. TPOB may contribute to expanding such inquiries on postmodernism beyond Pynchon in the future. Finally, the development of the TPOB also contributes to the growing number of Pynchon studies using digital humanities methods (Herman et al. 2003, Tsatsoulis 2013, Letzler 2016, Muth 2019a, 2019b, Hänggi 2020, Ketzan 2022).

Technical methods in creating the TPOB 1.0 release are presented at the end of this paper, and the following sections present formal descriptions and experiments to unearth new knowledge about Pynchon studies, demonstrate how TPOB metadata may highlight research gaps in Pynchon studies, and discuss how the analysis of Pynchon studies may be expanded to bibliometric studies of postmodernism, and perhaps further, in the future.

Overview

“Back down in the computer library, in storage, quiescent ones and zeros scattered among millions of others” — Vineland (1990, pg. 115).

Figure 1 presents the counts of metadata on more or less “scholarly”3 contributions in TPOB: 84 monographs, 1,234 articles, 915 book chapters and sections, and 300 reviews of Pynchon studies, for a total of 2,533 Pynchon studies, total. Not visualised here, TPOB additionally contains metadata on 451 reviews of Pynchon’s texts from a mix of scholarly sources and news media, as well as 178 texts which Ryckx classified as “cultural journal” reviews and articles about Pynchon. As the 1.0 update of TPOB focused on the academic sources of Pynchon studies, I have not extensively reviewed or updated these media entries. While I have updated and expanded upon Ryckx’s initial classification scheme, no particular claims are drawn from the taxonomy followed; the main aim of TPOB 1.0 is simply to have as many resources in one place as possible, with as much useful metadata as possible, and taxonomies can be further developed in the future if they are useful.

Figure 1

Count of Pynchon studies by format in TPOB 1.0.

Text-to-commentary ratio

The first observation from the data now available in Figure 1 is that this is simply a staggering amount of scholarship and discourse on a living author’s texts: 53,805 pages of more-or-less academic commentary on Pynchon. The page total of Pynchon’s novels plus Slow Learner is 4,225 going by first editions, so while book pages can have significantly varying numbers of words due to formatting, this results in roughly 12.7 pages of commentary for every page of fiction published by Pynchon. For another perspective, using word token counts that includes Pynchon’s uncollected short stories, Pynchon has published 2,177,584 words of fiction, which results in roughly 2.5 pages of commentary for every 100 words Pynchon has written.4 These metrics, even if approximate, could be dubbed text-to-commentary ratios.5

I hypothesize that by the metric of text-to-commentary ratio, Pynchon is among the most, if not the most studied living author in the English language. Testing this hypothesis would require similar bibliographic projects for the scholarly reception of Toni Morrison, Margaret Atwood, J. K. Rowling, and other obvious and non-obvious candidates, which is now more feasible than ever thanks to availability of data and the increasing ease of processing.6 As bibliographic metadata becomes more available and open, it may be possible to more quickly and accurately identify new “rising stars”, authors whose texts are suddenly receiving lots of academic attention, and conversely, trace the wane of once much-studied authors who no longer garner large volumes of criticism (supplemented with sales figures and whether books remain in print, for instance).

A modeling of text-to-commentary ratios in fiction inevitably implicates the notion of the canon, a vigorously contested discourse. As summarized by Jaume Aurell, “From the 1970s onwards, postmodern, postcolonial and feminist criticism has reactivated the disapproval of the canon. Its academic and pedagogical usefulness remains challenged by historians who have generally been skeptical of its critical worth, its role as an instrument of hierarchization and standard of quality” (2022). Could wider calculation of text-to-commentary ratios — of Pynchon and others — challenge any assumptions around canonicity? It is, at least, possible that new data may spark new conversations.

Bibliometric chronology

While I have added hundreds of entries to TPOB after the Vheissu.net ingest, this is still certainly not comprehensive; no doubt there are many more Pynchon studies, especially in non-English languages. Nonetheless, TPOB provides metadata on 2,533 more or less “academic” Pynchon studies, with chronological distribution presented in Figure 2.7 A limitation of this method is that, of course, various books and journals format pages differently, but presenting the bibliography by page count, rather than frequency of publications, offers a fresh perspective for analysis.

Figure 2

Pynchon studies by page counts and year of publication.

Figure 2 provides some evidence that the publications of new novels by Pynchon has invigorated scholarly output in the subsequent years; observe the renewed productivity form 1997 onward, following Mason & Dixon, then the years following 2009’s Inherent Vice. The declining pattern of publications in recent years is some evidence that the production of Pynchon studies in article and book chapter format has declined to low levels especially from 2021 on, although monographs remain a consistent source. Following the comprehensive summing up of Pynchon studies in three major edited volumes, Thomas Pynchon in Context (2018), Thomas Pynchon, Sex, and Gender (2018), and The New Pynchon Studies (2019), it may take more time for another major edited collection.

Is Pynchon studies in decline? Perhaps. This very journal launched as Orbit: Writing Around Pynchon in 2012 but broadened its scope to Orbit: A Journal of American Literature in 2016, possibly as Pynchon studies was no longer active enough to sustain a full journal. Pynchon’s surprise new 2025 novel, Shadow Ticket, will undoubtedly generate more studies, as will the new Vineland-inspired film, One Battle After Another, by director Paul Thomas Anderson, as well as the eventual availability of the Thomas Pynchon Archive to scholars by The Huntington Library. While the volume of Pynchon studies may have reduced in recent years, given a hypothesis that scholars are exhausting the interpretive gold mine of the texts, below I will illustrate how the TPOB may reveal research gaps in Pynchon studies that may yet be fruitful.

Most Frequent Words in Pynchon Studies Titles

“What’s the most frequent word?” asks Jessica. “Your number one.”

“The same as it’s always been at these affairs,” replies the statistician, as if everyone knew: “death.” — Gravity’s Rainbow (1973, pg. 32).

The dialogue above demonstrates that Pynchon has long been attuned to the quantitative aspects of language — “most frequent word” — well before the personal computer placed text query and myriad forms of natural language processing in the hands of any scholar. The most frequent words in Pynchon scholarly titles (monographs, articles, and book chapters/sections which have a title) are presented in Table 1, with words lemmatized to combine frequency of word forms and plurals, and stopwords (e.g. of, in, the) excluded.8 Unsurprisingly, the titles of Pynchon’s novels are frequent, with Gravity’s Rainbow most frequent, in keeping with broad consensus of its stature in Pynchon’s oeuvre. The high number of dedicated Vineland studies is a bit surprising, given its perennial reputation as a minor Pynchon novel, “hardly anyone’s favorite Thomas Pynchon novel”, per Peter Coviello (2020). The abundance of Vineland studies may be due to the timing of its publication; in 1990, Vineland was fresh material for a large and growing Pynchon Industry at the time, as Figure 2 above shows.

Table 1

Most frequent words as lemma in Pynchon studies titles.

Lemma	Frequency	Lemma	Frequency
pynchon	929	world	33
thomas	448	literature	33
gravity	317	war	29
rainbow	204	inherent	27
lot	158	science	26
crying	151	time	25
fiction	120	introduction	25
postmodern	116	contemporary	24
america	93	metaphor	24
novel	74	text	23
mason	73	note	23
history	72	oedipa	22
dixon	64	century	21
vineland	60	modern	21
read	58	life	19
day	47	technology	18
narrative	43	bleed	18
entropy	42	early	18
new	42	use	18
paranoia	39	space	18

Examining the most frequent words in Pynchon studies titles with Pynchon’s name, surname, and book titles excluded (Table 2), the words present a familiar consensus summary of Pynchon’s texts: postmodern American fiction about history, paranoia, war, and science.9 The 51 instances of entropy reflect the abundance of focus in earlier Pynchon studies on the concepts/term; David Cowart wrote that “The most abused of these critical ‘keys’ to Pynchon is the concept of entropy” (1982, pg. 2), while Brian McHale summarized the “topoi of Pynchon criticism” as “entropy, Preterition, Manichaean dualities and excluded middles” (1990, pg. 139). TPOB allows us to gather evidence of such trends in Pynchon studies from titles alone (Figure 3).

Table 2

Most frequent words as lemma in Pynchon studies titles, with titles of Pynchon’s texts excluded.

Lemma	Frequency	Lemma	Frequency
fiction	120	oedipa	22
postmodern	116	century	21
america	93	modern	21
novel	74	life	19
history	72	technology	18
read	58	early	18
narrative	43	use	18
entropy	42	space	18
new	42	source	18
paranoia	39	story	17
world	33	post	17
literature	33	system	17
war	29	quest	17
science	26	reader	16
time	25	dream	16
introduction	25	zone	16
contemporary	24	end	16
metaphor	24	death	16
text	23	self	15
note	23	work	15

While the per-year frequencies were never high, it seems that the “entropy” trend in Pynchon studies lamented by Cowart is now long over.

Figure 3

“Entropy” in the title of Pynchon studies.

It is notable that in the most frequent words of the Pynchon studies titles, only one of Pynchon’s fictional characters is relatively highly frequent: Oedipa, the protagonist of The Crying of Lot 49.10 Slothrop appears only 7 times in scholarly titles, Pierce Inverarity twice and Maxine only once. This lack of interpretive focus on the characters of Pynchon’s fiction aligns with a longstanding criticism of the thinness of Pynchon’s characters (see e.g. Locke 1973, Hume 1992, Logan 1998), although there is evidence from reception of more “fleshed out” and “believable” characters in Pynchon’s later novels (see e.g. Duyfhuizen 2007, Ketzan 2022, pg. 24–28). This evidence underscores that if one character from Pynchon’s hundreds has had a notable reception, it is Oedipa.

To dig into words in the titles further, I count 273 unique verb lemmas in the English language scholarly titles, from abandon to deconstruct to write.11 Verbs can provide evidence of not only the content of Pynchon studies but what actions these many Pynchon studies claim to be doing (Figure 4).

Figure 4

Most frequent verbs (as lemmas) in Pynchon studies titles. Max n = 24.

Like most contemporary literary studies, Pynchon studies aim to read (frequency: 24), know (10), teach (8), reread (5), write (7), remember (5) and so on, although even the most frequent verbs are quite low frequency, and thus not very representative of 2,000+ titles. Developing interpretations based on the verbs in these titles is additionally complicated by a common construction, “Quote from Pynchon text”: Scholarly Aim. Some examples:

“How pleasant to watch Nothing”: Narrativity and Desire in V. (Gilmore 2012, quoting V., pg. 528)

“Strung into the Apollonian Dream”: Pynchon’s Psychology of Engineers (Tabbi, 1992, quoting Gravity’s Rainbow, pg. 754)

“Chain of Links” or “Disorderly Tangle of Lines”?: Alternative Cartographies of Modernity in Thomas Pynchon’s Fiction (Cornis-Pope 2002, quoting Mason & Dixon, pg. 349)

The verb lemma go (frequency: 10) in the Pynchon studies titles is an example of a notable result, but inspection of the titles reveals that go is often found in a quotation from a Pynchon text, complicating generalization based on verbs:

“Going Round and Round with Old Gravity”: Technological Progress and Escape in Thomas Pynchon’s Gravity’s Rainbow (Dalsgaard, 2000, quoting Gravity’s Rainbow, pg. 361)

“He Could Go to Malta and Possibly End It”: Malta as “Prime Location” in The Epilogue of V. (Simonetti, 2015, quoting V., pg. 372)

Interpreting verbs alone in the Pynchon studies titles thus seems unfruitful, but as future work, such quotations from Pynchon’s texts could all be identified, their textual locations mapped, and patterns investigated. I speculate that even with 2.5 pages of scholarly commentary for every 100 words by Pynchon, scholarly commentary may cluster more around certain passages, sentences, and even words (as was the case with entropy). Closer analysis of the scholarly titles presents opportunities for intertextual interpretation and modeling, as well, as scholars will sometimes include a quotation from another writer within a Pynchon studies title. For example, “‘Something More Than a Rifle’: Firearms in and around Thomas Pynchon’s Mason & Dixon” (Rossi, 2014), begins with a quotation by author William T. Vollmann, not Pynchon. Mapping such quotations chosen by Pynchon studies titles would be one step toward modeling the specific textual locations (and intertextual links) which have received the most commentary, and also thus the least commentary, which could point scholars towards “research gap” passages in Pynchon’s texts which have so far gone unexplicated.

Languages of Pynchon studies

English is by far the largest represented language in Pynchon studies, with 1,799 studies in TPOB totaling 28,120 pages, but studies in many other languages are included as well and can no doubt be added to. Table 3 presents the languages of Pynchon studies articles, book chapters/sections, and monographs in TPOB.

A recent set of reviews in Orbit of Pynchon studies in Polish, Spanish, Japanese, and Italian highlighted many issues in the international scope of Pynchon studies, including “new insights that go beyond the existing anglophone canon of scholarship” (Chetwynd et al., 2021). Table 3 supports many of the arguments of Chetwynd et al.’s article, which also included an interview with Michel Ryckx on his process and aims for expanding Vheissu.net. The linguistic limitations of Pynchon studies certainly apply to literary studies more broadly; as Hammarfelt writes, “bibliometric studies of research fields in the humanities need to incorporate non-English […] publications in order to produce valid and fair results” (2012, 172). To improve non-English representation in TPOB, I have updated and added entries where reliable bibliographic metadata could be found. But certainly, TPOB 1.0 version can be greatly improved further in non-English language entries, especially in the areas discussed by Chetwynd and Ryckx, notably studies from Asia and Africa.

Table 3

Languages and page counts (where available) of scholarly titles in TPOB.

Language	Studies	Total Pages
English	1,799	28,120
German	66	1,377
French	68	1,083
Japanese	74	962
Italian	33	492
Polish	21	363
Korean	15	306
Spanish	17	254
Hungarian	6	78
Russian	6	70
Finnish	4	65
Portuguese	5	60
Farsi	1	20
Dutch	2	11
Mandarin Chinese	1	8
Swedish	1	8
Danish	1	6
Serbo-Croatian	1	6

Pynchon’s Intertextual Fields

“any text is an intertext; other texts are present in it, at varying levels, in more less recognisable forms: the texts of the previous and surrounding culture.” — Roland Barthes, “Theory of the Text” (1987, pg. 39).

Intertextuality was a term first proposed by Julia Kristeva in the 1960s, with the “literary word” as “an intersection of textual surfaces rather than a point (a fixed meaning), as a dialogue among several writings” (1980, pg. 65). The quantification of literary intertextuality is growing, via methods of detecting text re-use in primary texts (e.g. Büchler et al. 2014) and stylometric features (e.g. Coffee et al. 2012, Dexter et al. 2017, Manjavacasa et al. 2020). But the titles (and content) of scholarly commentary presents another avenue for considering intertextuality quantitatively, not strictly as “present” or interpretable in the primary texts (i.e. in Pynchon’s fiction) but present in the reception, in the intertextuality brought to the text by scholars.

In Pynchon studies, there are a number of contributions that name, besides Pynchon himself, a second author or second author’s text in the title, whether for an intertextual or comparative study or a theoretical lens, e.g.: “Gravity’s Rainbow and Mikhail Bakhtin’s Poetics of the Novel” (Morris, 1998); “Toshi no Meta-Fiction: Barthelme to Pynchon ni Miru Toshi” (Miyamoto, 1983); “Sexe et association dans Tristram Shandy et Gravity’s Rainbow” (Kinsley, 1986). Identifying the named authors and texts in the titles could support a substantial amount of analysis, but identifying them computationally and automatically from the title strings is not a simple task. While I explored methods for automatic named entity recognition to speed up the identification of these named authors, in the end, in favor of accuracy, I manually inspected and tallied the authors and texts explicitly named in the titles (Figure 5).

Figure 5

Frequency of authors or their texts named in the titles of Pynchon studies.

238 out of 2088 contributions (which have a title) name an author or text other than Pynchon, or ~11% of titles, and 173 authors (or their texts) are named in total: mostly canonical literary authors and philosophers whose names are familiar to Pynchon scholars. While 11% is a small sample of the whole of titles available, some observations may be made. It is interesting that, apart from Joyce, a large percentage of “Pynchon and X” studies involve Pynchon’s contemporaries: John Barth, Don DeLillo, Kurt Vonnegut, William Gaddis, suggesting that for contemporary authors, the “Authors 1 and 2” subgenre of academic studies may mostly involve comparative studies with other contemporaries. But moving beyond the word tokens of the titles requires the addition of more metadata.

To obtain more metadata on the authors named in Pynchon studies titles, using the rich online connected databases variously dubbed the Semantic Web or Linked Open Data, I first automatically obtained the WikiData ID number for each author using the MediaWiki and WikiData Python APIs.12 Almost all of the named authors already have WikiData entries, even, to my surprise, the relatively minor ones. With the WikiData ID number for each author, the wealth of metadata on WikiData available for each author was then queryable via code. After a small number of exclusions (e.g. the Brothers Grimm as two people, “Wanda Tinasky” as a pseudonym), some observations of the WikiData metadata can be made (sample in Table 4).13

Table 4

Sample of metadata obtainable through WikiData APIs for named persons in Pynchon studies titles.

	Name	WikiData_ID	Nationality	Year_of_birth	Place_of_birth
0	John Barth	Q315683	<wikidata.entity.Entity Q30 ‘United States’>	1930	<wikidata.entity.Entity Q755862 ‘Cambridge’>
1	Don DeLillo	Q310048	<wikidata.entity.Entity Q30 ‘United States’>	1936	<wikidata.entity.Entity Q18426 ‘The Bronx’>
2	James Joyce	Q6882	<wikidata.entity.Entity Q27 ‘!reland’>	1882	<wikidata.entity.Entity Q2528857 ‘Rathgar’>
3	Kurt Vonnegut	Q49074	<wikidata.entity.Entity Q30 ‘United States’>	1922	<wikidata.entity.Entity Q6346 ‘Indianapolis’>
4	William Gaddis	0456958	<wikidata.entity.Entity 030 ‘United States‘>	1922	<wikidata.entity.Entity Q60 ‘New York City’>
…	…	…	…	…	…
164	William Vollmann	Q29344	<wikidata.entity.Entity Q30 ‘United States’>	1959	<wikidata.entity.Entity Q65 ‘Los Angeles’>
165	William Wordsworth	Q45546	<wlkidata.entity.Entity Q174193 ‘United Kingdo…	1770	<wikidata.entity.Entity Q241684 Cockermouth’>
166	Wolfgang Hildesheimer	Q51510	<wikidata.entity.Entity Q193714 ‘Mandatory Pal…	1916	<wikidata.entity.Entity Q1055 ‘Hamburg’>
167	Wyndham Lewis	Q780102	<wikidata.entity.Entity Q145 ‘United Kingdom‘>	1882	<wikidata.entity.Entity Q470594 ‘Amherst’>
168	Yevgeny Zamyatin	Q204868	<wikidata.entity.Entity Q34266 ‘Russian Empire’>	1884	<wikidata.entity.Entity Q157447 ‘Lebedyan‘>

With quicker access to a range of metadata about the named authors in Pynchon studies titles, descriptive intertextual metadata can be quickly retrieved. For example, the average year of birth of the named authors is 1883 (or 1881 weighted for mentions). Regarding gender, 157 are male, 12 are female, with a complete list of the latter being: Hannah Arendt, Margaret Atwood, Djuna Barnes, Rachel Carson, Angela Carter, Joan Didion, Marlen Haushofer, Tama Janowitz, Elfriede Jelinek, Toni Morrison, Elizabeth Madox Roberts, Marilynne Robinson. With the linked metadata in WikiData, various inquiries extending from the TPOB could be developed in the future. As one experiment, I created a heat map based on the places of birth of all authors named in the titles of Pynchon studies (Figure 6).14

In the heat map can be observed the expected large cohorts born in North America and Europe, with other named writers born in Central America, India, Japan, and one each in South Africa and New Zealand. Maps are far from innocent interpretive vehicles, and while digital mapping technology has been critiqued as “an apparatus of logical positivism” and “naïve empiricism” (e.g. Leszczynski, 2014), we can recall that Pynchon was well aware of the limitations of geographic representation, most obviously in Mason & Dixon, but in Gravity’s Rainbow and Against the Day as well. Sascha Pöhlmann summarizes: “Mapping the world always means inventing and constructing it to some extent in Pynchon’s fiction, and the crucial issue is not whether the map is ‘true’ but what can be done with it” (2018, p. 69). And indeed this map is simply a model of birthplace metadata, and birthplace should not be conflated with any current national literature; George Orwell, for instance, was born in present-day India. Nonetheless, what “can be done with” this map is reveal literature gaps that future Pynchon studies could fill: the many fascinating studies just waiting to be written about Pynchon in relation to authors and texts in many other linguistic, national, and local contexts, perhaps not forgetting that “geography is as much spiritual as physical” (Against the Day, pg. 165).

Figure 6

Heat map of places of birth of authors named in Pynchon studies titles.

Stepping away from maps, the data on authors mentioned in Pynchon studies titles presents a simple model of what could be considered Pynchon’s intertextual field, a concept which has been discussed in analog literary studies (see e.g. Bazerman 2003, Berger 2014) but could be explored much more computationally through defined metrics that could be applied to any much-studied author. Certainly the creation of a full-text Pynchon studies corpus would provide data for a better model of an intertextual field than scholarly titles alone. Some intertextual fields could derive strictly from the text, beginning with those texts and authors unambiguously referenced in Pynchon’s fiction, such as Emily Dickinson or Rainer Maria Rilke. Next, some middle ground between reference and interpretation could be modeled; Nabokov is not named in Pynchon’s texts, but he has been interpreted as obliquely referenced in V. (Dugdale 1990, p. 181). Or, one could model the texts which commentators have broadly brought into dialogue with a target text, as I have here: Pynchon and Arendt, Pynchon and Joyce, Pynchon and DeLillo, etc.

Myriad methodological steps could be considered, as suggested by my quotation of Barthes at the beginning of this section, as Barthes presented intertextuality as an impossibly large and complex phenomenon: “the whole of language, anterior or contemporary, comes to the text, not following the path of a discoverable filiation or a willed imitation” (1987, p. 39). And intertextuality in Pynchon can certainly be manifold, as discussed by David Seed, who writes that Pynchon’s “fiction consists to an important extent of a montage of historical material garnered from diverse sources”, and also that “we should approach [Pynchon’s] texts as fields where different representational systems and verbal registers are constantly encountering one another” (2011, pg. 112). While the possibilities of modeling intertextuality quickly multiply, including not only “texts” but languages and representational systems, some model of intertextuality, supported by the TPOB and other bibliometric projects, would be promising for more research.

Ingest, Update, and Metadata Enrichment

“Back down in the computer library, in storage, quiescent ones and zeros scattered among millions of others […]” — Vineland (1990, pg. 115).

Thanks to Ryckx’s diligence in including the unique EAN (European Article Number) in bibliographic entries, almost all of the book-length bibliographic items in Vheissu.net were easily importable into Zotero, although in the case of book chapters, additional metadata had to be queried and added, aided by querying EANs on WorldCat. This left 1,532 academic articles in plain text on Vheissu.net to import into Zotero. The DOIs for 560 of them were automatically identified using CrossRef’s Simple Text Query online API.15 With the DOI, automatic ingest into Zotero captured the most important metadata, although surprisingly, page numbers often did not ingest into Zotero using DOIs, and I manually added page-counts for hundreds of articles.

I initially considered a rule-based Python script to convert the remaining 972 plain text article references into BibTex for Zotero import, but the method of large language model (LLM) proved quick, easy, and accurate for this task, demonstrating the usefulness of LLMs in bibliographic tasks. A number of current-generation LLMs automatically converted:

Brandt, Stefan L. “Polysemantik und Entropie: Zur Ästhetik der Ambiguität bei Henry James und Thomas Pynchon”. Literaturwissenschaftliches Jahrbuch der Görres-Gesellschaft Vol. 46 p. 303–329 (2005). [German]

to

@article{brandt2005polysemantik,

author = {Stefan L. Brandt},

title = {Polysemantik und Entropie: Zur Ästhetik der Ambiguität bei Henry James und Thomas Pynchon},

journal = {Literaturwissenschaftliches Jahrbuch der Görres-Gesellschaft},

volume = {46},

pages = {303–329},

year = {2005},

note = {German}

}

The LLM method thus made the ingest of the 972 articles into basic BibTeX quite simple.16 I similarly imported 299 of Vheissu.net’s “Review Essays and Reviews of Criticism” — reviews of Pynchon studies — as well as 178 “Cultural Journal Articles” from such general readership publications as The Atlantic and New Statesman, and 445 mostly journalistic reviews of Pynchon’s fiction. I have generally followed Vheissu’s classification schema, with some adaptations, for these many articles, but this could be revisited in the future.17 The articles imported using DOI naturally have more metadata, e.g. URLs, but basic bibliographic info is now ingested. Vheissu.net did not list the individual articles in the various Pynchon edited volumes; I have manually added entries for 426 of these book chapters from 30 edited volumes. Finally, additional new Pynchon studies, including books and articles from the last few years, were added.18

Conclusion and Future Work

I hope that TPOB 1.0 will be a useful resource for Pynchon studies. This paper has attempted to show how TPOB’s bibliographic metadata can both inspire and support new analyses in Pynchon studies, unearth research gaps, and lead to further bibliometric studies of postmodern literature and literary studies more broadly in the future. Some obvious further improvements to TPOB have been mentioned above: ingest more non-English sources, and manual check of more metadata. A more rigorous approach could be to complete a PRISMA 2020 metadata report for the systematic reviews of studies (Page et al. 2021).

Much more hypothesis- and data-driven research on the TPOB could be undertaken. Above all, the most ambitious would be to use TPOB to create a Pynchon scholarly corpus, containing as many full texts as feasible, which would give greater weight to the examination of intertextual fields or text re-use, i.e. which of Pynchon’s quotations are cited in the literature. But more could be done with the titles and metadata in TPOB already. I have not attempted to quantify the citations metrics of the Pynchon studies, a common inquiry in bibliometrics, as existing bibliometric databases do not contain a great number of Pynchon studies that fall outside the most established journals and publishers, but it might be worth knowing which studies have had the most quantifiable impact.

Back in 1993, in the pages of Pynchon Notes, Stuart Moulthrop and John McDaid imagined future hypertext models for “hypertextual application for literary studies”:

The most elementary would be a compendium of textual resources […]. A Gravity’s Rainbow web would include the primary text, any draft and source material that could be discovered, a comprehensive gathering of scholarship on the novel, bibliographic apparatus, and at least a sampling of related or implicated texts—which in this case would dwarf the other parts of the project. All these elements would be deployed within a framework that included specific links, facilities for developing new connections, and a “navigation” scheme for representing the structure of the information base and allowing users to move within it. The web would be designed to provide scholars engaged in research on Gravity’s Rainbow with a collection of existing work and with a tool for organizing and synthesizing that large body of resources. We might think of this model as an industrial application of hypertext—in the sense of “the Shakespeare industry,” “the Joyce industry” or “the publishing industry.” (141–42)

Or the Pynchon Industry, which TPOB presents our most comprehensive model of to date. While Moulthrop and McDaid continued their article by attempting to complicate future visions of a Gravity’s Rainbow hypertext into more radical experiments, now, over thirty years later, TPOB is a next, even if “elementary” step.

While even at 2000+ Pynchon studies in TPOB, manual methods and correction were often feasible for the present exploratory paper, future studies will certainly require more accurate automatic methods to scale the inquiries (for instance in named entity recognition), either if a Pynchon scholarly corpus can be built or research questions extended beyond Pynchon (e.g. to postmodernism studies). Lacunae and errors no doubt exist in TPOB 1.0, and while this is acceptable for the current resource location and exploratory aims, more rigorous checks will be required in the future.

The bibliography that TPOB ingested was only part of the content on Vheissu.net, which published and hosted a variety of materials beyond the bibliography, including archives of articles on Pynchon and original contributions by Ryckx and his collaborators, the largest of which was a voluminous index by Ryckx of thousands of named entities and terms in Pynchon’s texts, classified into his own ontology and accompanied by page numbers and related terms. To honor Ryckx’s wish to preserve Vheissu.net for future research, I plan to separately archive the website’s files in an open access repository after performing a copyright review to exclude copyrighted materials (such as newspaper articles on Pynchon) which Vheissu.net hosted.

Finally, I plan to maintain the TPOB with periodic updates and improvements, and I welcome contributors, most especially those with expertise in non-English Pynchon scholarship.

Notes

1. https://www.zotero.org/support/credits_and_acknowledgments. [^]
2. https://web.archive.org/web/20220323201738/http://www.vheissu.net/biblio/. [^]
3. Certainly a taxonomy of “scholarly” vs. “amateur”, “media”, or any other classification is porous and complex, and I will not attempt it here. [^]
4. Different software calculates the number of word tokens in a text with different methods, resulting in somewhat variable numbers. For this calculation, I used the tokenizer in Textometrie (TXM) 0.8.3. See Heiden 2010. [^]
5. I cannot find such a similar metric discussed in the bibliometric literature but would be unsurprised if it has already been suggested. [^]
6. For a visualization of frequencies of dedicated studies on numerous much-studied authors, according to the Modern Language Association International Bibliography, see Porter 2018. [^]
7. This figure includes page counts for about 96% of the academic publications in TPOB; the remaining 4% either had no page counts, are online-only publications, or page counts could not be located. [^]
8. Lemmatization by spAcy, using en_core_web_trf-3.8.0, https://spacy.io/models/en. [^]
9. Tables 1 and 2 present data from English-language titles only. The query results for entropy and character names which follow includes all languages. [^]
10. If Mason and Dixon as characters are excluded, as contained in a novel title. [^]
11. Again, this includes monographs, book chapters, and academic articles. Verbs were obtained using this part of speech tagger in spaCy, with the model same as above, footnote 5, then checked manually. For an exemplar of reading verbs in literary titles, see Moretti (2013). [^]
12. MediaWiki API 0.7.5 wrapper in Python, https://github.com/barrust/mediawiki. WikiData 0.8.1 client library, https://pypi.org/project/Wikidata/. [^]
13. WikiData categories queried based on WikiData ID, using the packages in footnote 12, above, included Nationality (P27), Year of Birth (P569), Place of Birth (P19). [^]
14. Longitude and latitude for the birth places were extracted by WikiData using the packages in the footnotes above, then visualized using folium 0.19.4, HeatMap plugin, https://pypi.org/project/folium/. [^]
15. https://apps.crossref.org/SimpleTextQuery. [^]
16. Claude 3.5 Sonnet was used for the task, but I observed equal performance by Gemini 1.5 and GPT-4o. [^]
17. Vheissu.net’s entries on 11 conferences and 270 presentations have not been ingested, and tracking presentations relating to Pynchon globally seems an unsustainable endeavor, in the absence of large resources that may not be feasible for such a project. These entries are archived at: https://web.archive.org/web/20220323201738/http://www.vheissu.net/biblio/. [^]
18. Recent publications queried at Scopus, Semantic Scholar, and Google Scholar. [^]

Acknowledgments

Thanks above all to Michel Ryckx. Thanks also to Tore Rye Andersen, Martin Paul Eve, and Jeff Severs for correspondence, and Zofia Kolbuszewska and Sascha Pöhlmann for sending bibliographic information on some hard-to-find books. Project documentation and data are available at: https://github.com/erikannotations/TPOB

Competing Interests

The author has no competing interests to declare.

References

Aurell, Jaume. 2022. “The Canon in History.” Rethinking History 26 (4): 439–65. http://doi.org/10.1080/13642529.2022.2135814.

Barthes, Roland. 1987. “Theory of the Text.” In Untying the Text: A Post-Structuralist Reader, edited by Robert J. C. Young, repr, 30–47. Boston: Routledge & Kegan Paul.

Bazerman, Charles. 2003. “Intertextuality: How Texts Rely on Other Texts.” In What Writing Does and How It Does It, edited by Charles Bazerman and Paul Prior, 0 ed., 83–96. Routledge. http://doi.org/10.4324/9781410609526.

Berger, Yitzhak. 2014. “The David–Benjaminite Conflict and the Intertextual Field of Psalm 7.” Journal for the Study of the Old Testament 38 (3): 279–96. http://doi.org/10.1177/0309089214527206.

Bredahl, Laura. 2022a. “Chapter 2. Bibliometric Tools for Academia.” Library Technology Reports 58 (8): 12–21.

Bredahl, Laura. 2022b. “Chapter 3. Applications of Bibliometrics.” Library Technology Reports 58 (8): 22–31.

Büchler, Marco, Philip R. Burns, Martin Müller, Emily Franzini, and Greta Franzini. 2014. “Towards a Historical Text Re-Use Detection.” In Text Mining, edited by Chris Biemann and Alexander Mehler, 221–38. Theory and Applications of Natural Language Processing. Cham: Springer International Publishing. http://doi.org/10.1007/978-3-319-12655-5_11.

Chetwynd, Ali, Andrea Brondino, Dominika Bugno-Narecka, Kodai Abe, Michel Ryckx, Romina Kipouridou, and Vit Vanicek. 2021. “Book Reviews, Special Pynchon-Scholarship-in-Languages-Other-than-English Edition, 2020.” Orbit: A Journal of American Literature 9 (1): 1–40. http://doi.org/10.16995/orbit.3404.

Chetwynd, Ali, Joanna Freer, and Georgios Maragos, eds. 2018. Thomas Pynchon, Sex, and Gender. Athens: The University of Georgia Press.

Coffee, Neil, Jean-Pierre Koenig, Shakthi Poornima, Roelant Ossewaarde, Christopher Forstall, and Sarah Jacobson. 2012. “Intertextuality in the Digital Age.” Transactions of the American Philological Association 142 (2): 383–422. [http://d