robin_anne_reid: (Dragon)
robin_anne_reid ([personal profile] robin_anne_reid) wrote2012-03-27 03:20 pm

Grant Writing and my Tolkien Corpus Project

I am teaching an undergraduate grant writing course for the second time, and will (perhaps! depending on scheduling issues) be teaching a grant writing course on the graduate level (humanities, social sciences, and arts only) next fall. One of the discussions that comes up a lot with my students is the frustration many of the best students have at not getting it "right" the first time. I talk about the process (and my grading is based on a modified portfolio version where the first drafts, worth very little in point value, are given 100% for effort, and lots of feedback for revision--multiple revisions). I am planning on sharing the link to this post with my current (and perhaps future students), because it concerns a grant project that I have been working on for some time: a Tolkien Stylistics Corpus project.

I might have just posted in the class, as I've done before with work, but I think there is so much mystery about the grant-writing process in general that the materials here might be of interest to other humanities scholars who want to work on grants.

I include information on the NEH grant categories I wrote for; my grant narrative; some screenshots of what the data looks like in the UAM Corpus Tool which I'm using; and the reader reports--direct from the NEH (they are anonymous of course) that explain the reasons my first draft (and by "first draft" here I mean "first draft submitted" not first draft written--I probably wrote about six drafts along the way--I'm not as careful as I used to be about saving each distinct draft with a number) were not funded. One of my linguist colleagues who is working with me on other grants was surprised at the tone of the comments--apparently linguist evaluators are nicer! I've been hearing since 1965 how trashy and popular and bad Tolkien is, and since the early 1990s how crappy my scholarship is for dealing with science fiction, fantasy, etc. that I'm more or less immune to it, so don't mind sharing. Additionally, in between some of the people who clearly think Tolkien OR stylistics OR both are worthless are some excellent responses that give me a lot of ways of re-conceptualizing and re-working the project (I always tend to take on TOO MUCH) over the next few years.

I like that the NEH sends the reader reports out on request--though odds are the review committee for later grants will be different, there are useful suggestions here that will apply no matter the make-up of the committee.

So, without further ado, my Tolkien Corpus Grant Materials!

The two grant programs are: First: Fellowships: This grant is the ultimate humanity grant, supporting up to a year of full-time scholarship. As you can imagine, the competition is fairly stiff. The information I received from the NEH (when I requested the reader report/evaluations, which the NEH will send out to people) is that: "The Endowment received 1,339 applications; the budget allowed for 80 awards." About a 6% success rate.

Second: This grant is the summer stipend, i.e. the Fellowship support for a summer of research. I was told that "The Endowment received 992 applications. Given the available resources, however, we could offer only 84 awards." So, 8.5% success rate.

Some sample Fellowship awards and A sample summer stipend award

My project deals with a text that embodies a foundational literary problem concerning canon formation. The problem is the fact that a beloved work which has influenced creative productions for decades can be mostly ignored by literature departments. Specifically, the fiction of J. R. R. Tolkien is often dismissed by scholars in the U.K. and the U.S. (notably Harold Bloom) as badly written escapism. This academic dismissal stands in sharp contrast to the respect of medieval scholars for Tolkien's scholarship on Beowulf and other medieval texts.

The argument which underlies my Tolkien Corpus project is that The Lord of the Rings and the rest of Tolkien's Legendarium (his fiction and poetry dealing with Middle-earth) are not bad modernist literature but excellent postmodernist texts. My argument can best be supported by stylistics. Stylistics applies linguistic principles to literary texts; corpus stylistics makes it possible to work with large amounts of text in an electronic format and to generate empirical data. The first step for my five-year project is creating a literary corpus. The stylistic scholarship which will result from this project will challenge widespread attitudes about Tolkien's writing style. I build on existing stylistic and rhetorical scholarship by major Tolkienists (Tom Shippey, Verlyn Flieger, Michael D. C. Drout) but apply linguistic methodology and digital technology to a larger amount of text to develop a stylistic analysis on a level that has not been hitherto been seen in Tolkien Studies.

J. R. R. Tolkien's The Lord of the Rings, published in the 1950s, has been enormously popular. Its success contributes to the controversy of its literary status since some critics refuse to concede the possibility of any aesthetic or literary merit in popular works. Initial critical response to The Lord of the Rings was mixed. The harshest critics are those trained in the modernist aesthetic; they judge the novel against criteria developed for the genre of realism and find its medieval romantic elements confusing, distasteful, or simply childish. The novel's association with The Hobbit, categorized as children's literature, and the persistent belief, developed during the Victorian and Edwardian periods, that "fairy-tales" are for children are responsible, in part, for the categorization of The Lord of the Rings as a simple children's tale about good and evil. Yet even at the time of original publication, critics such as C.S. Lewis and W.H. Auden saw something more in Tolkien's work, as their praise shows.

In the latter half of the 20th century, medievalists (Flieger, Shippey, and Jane Chance) argued the need to understand the influence of medieval source texts on the novel. However, analysis by medievalists has had little impact on the modernist canon, and the controversy over the literary merit of Tolkien's work continued into the 21st century. Although scholars in myth, folklore, religious studies, medieval studies, and linguistics have published widely on Tolkien, modernists, as a rule, have not. Critics specializing in 20th-century literature have ignored Tolkien or dismissed his work as "medieval manqué" created by a writer who retreated from the modern world into an imaginary past. In recent years, the question of Tolkien's work as postmodern has begun to be explored by a few scholars (Gergely Nagy, Flieger, Ralph Wood), but relatively little work has been done to develop postmodern readings of the Legendarium, nor have major postmodernist literary critics considered Tolkien's work.

There are signs prevailing attitudes may be starting to shift: Mfs: Modern Fiction Studies, a well-respected literary journal focusing on canonical and emergent texts, devoted their Winter 2004 issue to Tolkien's work. This issue shows that the vanguard of academic scholarship is prepared to move beyond a dismissal of Tolkien's work as "popular culture" or "trivialliteratur" (light fiction) and to contemplate its place in categories such as "modern fiction" or "literature." Michael D. C. Drout and Hilary Wynne, authors of "Tom Shippey's J. R. R. Tolkien: Author of the Century and a Look back at Tolkien Criticism since 1982" (2001), argue that the "biggest failing in Tolkien criticism…is its lack of discussion of Tolkien's style, his sentence-level writing, his word choice and syntax'' (123). As Drout has argued in "Tolkien's Prose Style and its Literary and Rhetorical Effects" (2004), the result of Tolkienists ignoring the question of the critical assumption that Tolkien's writing style is poor, i.e. not consistent with the criteria of realistic novels, "has had the unfortunate effect of ceding important ground to Tolkien's detractors, who, with simple, unanalyzed quotations, point to some word or turn of phrase and, in essence, sniff that such is not the stuff of good literature'' (137). A primary goal of my project is to begin to take back that ground, to challenge the too-easy assumptions of such criticism.

In order to fill the gap in Tolkien stylistics, I draw on Roger Fowler's Linguistic Criticism (1996) and a M. A. K. Halliday's An Introduction to Functional Grammar (1994). Fowler argues the problem with most attempts to analyze literary style is the lack of a consistent methodology. Linguistics provides a consistent, external methodology. Linguists tend to approach language descriptively rather than prescriptively, viewing language as a set of systems. Halliday's functional grammar offers a model which can be applied, word by word, clause by clause, to generate a quantitative analysis of a text. Applying a linguistic methodology to Tolkien's work serves the dual purpose of interpreting his work by a method seldom used and providing quantitative evidence to support an argument for the existence of a range of aesthetic effects and discursive elements in the work.

Existing stylistic or applied linguistic scholarship on Tolkien's novel in literary studies is limited; more work has been done on Tolkien's invented languages. The major literary scholarship consists of monographs by Verlyn Flieger, Brian Rosebury, and Tom Shippey. A handful of articles have also been published. One, the probable the basis for many dismissals of Tolkien's style, is Burton Raffell's 1968 essay, "The Lord of the Rings as Literature" in which he argues that the novel is not literature based on his comparison of a few passages from The Fellowship of the Ring to excerpts from D.H. Lawrence which are labeled by Raffell as "literature" without any justification than, apparently, his personal preference. More relevant work on Tolkien's style has been done by Elizabeth Kirk and Drout. The 1971 essay by Elizabeth Kirk, "'I Would Rather Have Written in Elvish': Language, Fiction, and 'The Lord of the Rings,''' argues against the assumption that a modernist aesthetic prioritizing individualistic and particular styles is universally good. Tolkien's choice to set disparate styles and registers against each other (the modern diction of the Hobbits, the archaic language of Rohan and Gondor) creates the sense of what Kirk calls a "communal consciousness," a style Kirk claims is "as different from the epic as it is from the novel'' (10). The only scholar besides myself to use Halliday in an analysis of Tolkien's work is Drout whose stylistic analysis focuses on two passages (Éowyn's fight against the Nazgul and Denethor's self-chosen death) and explores the intertextual parallels with Shakespeare. Both Drout's work and mine analyze only brief excerpts, a few hundred or thousand words. Analysis of greater scope requires a corpus, and digital tools.

My eventual goal is to prove the extent to which Tolkien's The Lord of the Rings reflects a postmodern aesthetic by means of stylistic analysis. At the most basic level, in Tolkien's created universe, reality is patently constructed by language. Instead of proceeding from the modernist assumption that language merely reflects reality, Tolkien began from the premise that the language of a people structures their culture. He created more than a dozen new languages in order to compose the mythology of Middle Earth which grounds The Lord of the Rings. Furthermore, The Lord of the Rings is abundantly self-referential and inter-textual with Tolkien's other fiction as well as with the worlds of early Germanic mythology and religion and the world of modern fairy-tales and folk-tales. Both the self-referential nature of The Lord of the Rings and its heavy dependence on language are often commented upon, although rarely explicitly connected with postmodernism. A characteristic less noted but still telling is the novel's heteroglossic nature. The Lord of the Rings cannot be easily categorized as either epic, or saga, fantasy, mythology, romance, or boys' adventure story (all applied to the work by genre scholarship) because Tolkien utilizes such a wide variety of discourses. Substantive proof of these claims can only be supplied by a wide-ranging, quantitative methodology which can encompass the 1200 pages (plus appendices) that make up The Lord of the Rings and which can then go on to consider his other fictions which are increasingly the focus of Tolkien studies.

Methods and Work Plan: This project is scheduled to take five years, with the final result being one or more monographs offering a comprehensive stylistic analysis of Tolkien's major fictions. The first year, which will be funded by the Fellowship, will be spent creating a Tolkien Corpus consisting of annotated texts of his major fictional works (The Hobbit, The Lord of the Rings, The Silmarillion, and his major short fiction and poetry). At least one article will be completed in the first year as well, an analysis of deictics (spatial indicators) in the novel.

This project faces a unique challenge: while it is possible other Tolkien corpora at various stages exist, the current copyright and intellectual property (IP) laws do not allow circulation of a corpus. I am working with my university's IP lawyer to gain permission from the Tolkien estate to create a corpus which will be stored on an external hard drive with a single back up (a different external hard drive) locked in my department office. The material will never be put on a computer, stand alone or networked, and the corpus will never be shared. I believe that the limits described constitute fair use of the work since it is for a scholarly purpose. Since creating a corpus requires hand-scanning thousands of pages of text, the creation of the corpus and initial levels of tagging will take the greater part of the first year. Additional levels of tagging will occur in later years. Saving the scanned pages as plain text files and then uploading them into the free program I will be using for annotation will take relatively little time and can be done immediately after a section has been scanned.

I will be using the UAM Corpus Tool, a free program created by a linguist, which be customized for various types of linguistic or stylistic analysis. My previous stylistics work required highlighting typed or photocopied text by hand to show different elements by color-coding, and then counting the results; the UAM Tool allows for complex and multiple layers of annotation that produce quantitative results that can be analyzed for significance ( .

The next major task will be annotating uploaded text files, marking, or tagging, distinct elements of the text in .xml. I estimate based on tests I have done that it will take 5-10 hours to mark, or tag, one layer, and an average of 40-60 hours to mark the most important grammatical elements, depending on the length of the chapter. The marking is done in layers, i.e. one layer will consist of marking and identifying all the clauses in a unit of text; another will be identifying categories of verbs. Halliday's grammar has six categories of processes (verbs). A decade ago, I read photocopies with a highlighter in hand, marking important elements; with the new programs, I still read through the text, highlighting and (most importantly) annotating, but the results are statistical tables generated by program; an example from a previous project can be seen in Appendix A. Annotation can only be done to a plain text file; the final version, viewed through the UAM window, is an unformatted chunk of text with green highlighting, not enjoyable to read for pleasure, but able to be analyzed statistically (examples in Appendix A).

Competencies, Skills, and Access: Since 2007, I have completed three (and published two; the third has been accepted by an editor of an anthology) stylistic analyses of Tolkien's work. These articles have mapped out a number of questions for my corpus stylistic project: for example, the question of how religion and myth operate in the text; the question of how male bodies are constructed as well as female bodies; the question of the variety of styles Tolkien worked in not only in his other published work but in earlier drafts, leading to my major claim concerning postmodernity in Tolkien. I have been co-teaching Tolkien's work since 2004 with Dr. Judy Ann Ford (a medieval historian) in both undergraduate and graduate courses on Tolkien. We have collaborated on several publications, including an essay on our online Tolkien graduate seminar that has been accepted for publication in the MLA anthology on Teaching Tolkien (ed. Leslie Donovan, forthcoming in 2012). Dr. Ford and I co-directed two summer NEH Institutes (2004, 2009) on Tolkien for school teachers. I have integrated stylistics work on Tolkien in all the classes and Institutes. I teach a graduate seminar on Stylistics and have integrated Tolkien's work into the class. I have served as the organizer of the Tolkien At Kalamazoo group for five years.

Final Product and Dissemination: Corpora are usually published online, but the typical corpus does not contain copyrighted material. A Tolkien corpus could not be published online. However, I plan to create a blog that covers the process of creating the corpus and discusses the process and developing scholarship. I have a current blog at <> that covers current corpus work I’m doing. A Corpus Stylistics Blog would be of interest not only to academics working in similar areas (Tolkien Studies and Stylistics) but to teachers and to fans. Academic publication in peer reviewed journals and an academic press would also be sought.

Copies of the panelists' ratings and written evaluations of your proposal are included with this memorandum. The range of possible ratings is Excellent (E), Very Good (VG), Good (G), Some Merit (SM), and Not Competitive (NC). Please keep in mind that panels are one stage of NEH review. The panelists provided both initial comments on each application before their panel meeting and final comments after discussion of each application during the meeting. Panelists' opinions and ratings may have changed in the course of the deliberations.

The panelists' names and references to other panelists or applicants have been omitted. Additional excisions reflect the Endowment's policy to hold in confidence the contents of letters of recommendation.

Panelist 1

This may be a worthy project in some respects, and its author understands and appreciates Tolkien’s works, but as presently conceived it has several limitations.

First, the idea that Tolkien’s works are good post-modernist literature is intriguing and probably correct, but follows in the wake of other claims that the literature of fantasy and other forms of non-realist writing are appropriately appreciated through the lens of post-modernist critics such as Deleuze and Guattari, etc. As popular literature is increasingly studied, the case for examining hitherto non-canonical literature seems more self-evident.

More important, the writer does not clarify sufficiently what kinds of stylistic features she expects to find, or their significance. I question, in any case, whether computer-generated data applying a quantitative analysis of the functional grammar of a text is the best place to go for subtle and analytical understandings of narrative prose, or that “the existence of a range of aesthetic effects and discursive elements in the work” is a sufficiently specific conclusion.

At the least, this seems an indirect and labor-intensive way of approaching topics which could be studied through close stylistic analysis. The sample page which the applicant provides convinces me that she will be able to reach some conclusions, but that these may not relate very centrally to the issue of whether we should admire Tolkien’s creative works.

And finally, this is a multi-year project in an early stage. It would certainly not be completed within the grant period. In the first year she proposes to complete “a Tolkien Corpus consisting of annotated texts of his major fictional works,” so she would not have begun the stylistic analysis by the end of the grant period. Moreover, if she has not already done so, she may underestimate the time and effort in “hand-scanning thousands of pages . . . and initial levels of tagging. . . . [and] Additional levels of tagging [which] will occur in later years.” The actual results in projected publications also need to be worked out more fully.

If the writer wishes to reapply, she should be encouraged to frame a clearer set of questions and possible results based on computer data, or to expand her notion of stylistics so that the computer portion of her project is more specific and perhaps less central to the outcome.

Preliminary Rating: NC: Not Competitive
Additional comments after panel discussion:
Final Rating: NC: Not Competitive

Panelist 2

The technical aspects of the proposal seem to be in line with what is required to establish electronic corpora, but the work plan for a five-year project is not suitable for this competition. The feasibility is doubtful at this stage and the material to be completed during the grant period is not specified.
Preliminary Rating: SM: Some Merit
Additional comments after panel discussion:
Final Rating: SM: Some Merit

Panelist 3

I do not see the value of rescuing Tolkien for scholarly or pedagogical use through computational stylistic analysis. If Tolkien is valuable, Reid and other scholars should continue researching his work (and they do; Tolkien research may not be on a par with Shakespeare, but he is not totally excluded from the field). More to the point: rescuing Tolkien by performing a statistical analysis of his style is misguided. Style is impossible, utterly impossible, to quantify in the manner proposed here. This form of analysis may be acceptable to linguists, but it is totally inappropriate for literary studies, which demands close attention to detail--to individual passages--not to accumulated data about the frequency of words or semantic units.
Preliminary Rating: NC: Not Competitive
Additional comments after panel discussion:
Final Rating: NC: Not Competitive

Panelist 4

This project seems to be a promising one that could one day receive NEH funding. The proposal is professional and well articulated. The candidate has the relevant experience and publications to make this seem viable. The project promises to lay the groundwork for future scholars, including the candidate herself. Nevertheless, there are some difficulties with funding this project at the moment. Most notably, its very viability now appears to hinge upon whether the Tolkien estate grants permission for the work to be done. This is no small barrier. The NEH cannot give upwards of $50,000 toward a project without the certainty that it can legally proceed. Secondly, while this proposal makes a powerful claim that Tolkien’s Middle-earth writings are “not bad modernist literature but excellent postmodernist literature,” it needs to do more to show how this statistical analysis will (or could possibly) quantify what is modernist as opposed to what is postmodernist. That link is difficult to make but crucial again to proving the viability of the work; some detailed examples could be helpful.
Preliminary Rating: G: Good
Additional comments after panel discussion:
Final Rating: G: Good

Panelist 5

1-Reid’s project “Tolkien Corpus Project” proposes to counter the view that Tolkien’s “Legendarium . . . are not bad modernist literature but excellent postmodernist texts.” She intends to do so by applying “linguistic principles to literary texts, [and to gather empirical data to support her argument]; corpus stylistics makes it possible to work with large amounts of text in an electronic format and to generate empirical data.” This will certainly generate a great deal of information about Tolkein’s use of language, to create his fictional language[s].

No negative aesthetic response can be countered and overturned by mere data; hopefully the information will lead to an understanding of a Tolkein aesthetic.

2-Reid has been engaged in has published the results of some of this research that have been apparently well received.

3-Reids plans are long range (five years) “offering a comprehensive stylistic analysis of Tolkein’s major fictions. . . . The first year, which will be funded by the Fellowship, will be spent creating a Tolkien Corpus consisting of annotated texts [nature of annotations not specified: inter-textual? extra-literary? Etc.] of his major fictional works (The Hobbit, The Lord of the Rings, The Silmarillion, and his major short fiction and poetry). At least one article will be completed in the first year as well, an analysis of deictics (spatial indicators) in the novel.” This seems too ambitious enough.

4- I think there is more than one project at hand here. The annotation of the major fictional works and short fiction and poetry is already quite an undertaking, which is barely addressed, though it is intended to be the immediate focus of the grant year. The linguistic analysis is another. The various elements of this very large and complex project need to be discreetly separated, and individually submitted for funding.

5. Reid is undertaking a lifetime project. Let her begin.

Preliminary Rating: NC: Not Competitive
Additional comments after panel discussion:
Final Rating: NC: Not Competitive

As with all applications submitted to the NEH, your proposal was read and discussed by knowledgeable persons outside the agency, who advised the Endowment about its merits. The NEH staff commented on matters of fact or on significant issues that otherwise would have been missing from these evaluations and made recommendations to the National Council on the Humanities. The National Council meets at various times during the year to advise the NEH chairman on grants. The Chairman took into account the advice provided during the review process and made all funding decisions, as is prescribed by law.

Copies of the panelists' ratings and written evaluations of your proposal are included below. The range of possible ratings is Excellent (E), Very Good (VG), Some Merit (SM), and Not Competitive (NC); (REC) indicates a panelist Recusal. Please keep in mind that panels are the first stage of NEH review and that the panelists sent us their evaluations and comments online.

The panelists' names and references to other panelists or applicants have been omitted. Additional excisions reflect the Endowment's policy to hold in confidence the contents of letters of recommendation.

The application deadline for the next NEH Summer Stipends competition is September 27, 2012. Application guidelines will be posted on the NEH website in May 2012.

Panelist 1

This project has some merit, but, while it aims to mount a defense of Tolkien's corpus on literary--that is, stylistic--grounds, its planned method and argument sidestep the question of literary value. Its focus on proving that similar stylistic constructions are to be found in Tolkien and in the great canonical writers is fine as far as it goes, but one could conduct the same exercise on any pair of texts without taking account of, or illuminating, their relative literary value. One could calculate, for example, that chimps with typewriters could collaborate on a work that, pursued over a billion years or so, would contain the entire works of Shakespeare; but if that work were also to contain much else, would it be Shakespeare? or comparable to Shakespeare?

Rating: SM

Panelist 2

Robin Reid's proposed "Tolkien Corpus Project," a computer-aided stylistic analysis of Tolkien's corpus, possesses a certain appeal for its thoroughness and rigor. The author's interest in defending Tolkien's work, on stylistic and rhetorical grounds, from its detractors also is understandable in the light of Tolkien's mixed critical reception (however popular his work may be with readers). The project strikes me as solid and thoughtfully laid out. That said, a five year project whose goal is to demonstrate that Tolkien's Lord of the Rings and the rest of his Legendarium "are not bad modernist literature but excellent postmodernist texts" does not inspire confidence. Not only are these two terms highly contested ones but there is little in the proposal to suggest that the author has anything approaching a nuanced grasp of the distinctions between and interrelations of modernist and postmodernist fiction.

Rating: SM

Panelist 3

Interesting attempt to redefine Tolkien as a proto-postmodernist, rather than a "merely" popular writer. However, it is not clear why an electronic "corpus" is required in order to conduct the stylistic analysis the applicant proposes (why can't existing e-books be used? concordances of Shakespeare and the Bible were compiled from hard copies of much larger oeuvres), nor, if the project were to be funded, how could the copyright issues be resolved. It's also not clear whether such an analysis would be likely to rehabilitate Tolkien's reputation.

Rating: SM

Panelist 4

Tolkien is perennially popular on college campuses among student readers if not on syllabi. Reid's plan to use stylistics sounds promising for quantifying Tolkien's aesthetic and rhetorical moves. Her extensive teaching and publication experience with Tolkien demonstrate her quality as the interpreter of this project, and the previous NEH seminars for high school teachers on Lord of the Rings show her longstanding dedication to the material. As she notes, this project is enormous and will take many years, far beyond the scope of the Summer Stipend. She is also forthright about a significant concern: not being able to circulate the corpus she creates. While discussing this matter on her blog will bring it out into the open, I can't quite determine whether that action will help other Tolkien scholars in their own analysis.

Rating: G

Post a comment in response:

Anonymous (will be screened)
OpenID (will be screened if not validated)
Identity URL: 
Account name:
If you don't have an account you can create one now.
HTML doesn't work in the subject.


Notice: This account is set to log the IP addresses of everyone who comments.
Links will be displayed as unclickable URLs to help prevent spam.