The Basic Outline of a Science Article

With many thanks to my collaborator Katie Kinnaird — I will have to post the Twitter account and link later from which this comes.

  1. Introduction: Tell me what you’re going to tell me.
  2. Literature Review: Tell me what everyone else told you, and then in Current Study tell me what they’re missing.
  3. Methodology: In Data tell me how you got the data and in Analysis tell me what you are going to do to the data.
  4. Results: Tell me what the data told you.
  5. Discussion: Tell me again what the data said but in a way that anyone could understand, and then tell me how it’s relevant and how we can apply the findings.
  6. Conclusions: Tell me what you told me.
  7. Limitations: Tell me if you had any issue and why they are either/both not your fault and/or trivial enough that this is all still publishable.

It’s fairly reminiscent of the DoD’s Cue — View — Review methodology for instruction.

In the Balance

Henry Glassie once observed that there were two great traditions of scholarship in folklore studies, one oriented toward data and the other toward theory. In the one oriented toward data, the analyst pieces together what theory she needs in order to explain the data at hand. Done well, such studies, Glassie noted, often offered data in excess of theoretical explanation, leaving the door open to future analyses by other analysts with different theories. In the tradition of scholarship oriented toward theory, the analyst begins with a theoretical construct and seeks out data to affirm it, revise it, refuse it.

Neither tradition is better than the other, and, in all honesty these aren’t separate traditions as two poles within the domain of folklore studies, though this axis of attention surely exists in other domains as well. At least in the American tradition(s), there are “no ideas but in things.” On the whole, we tend to look somewhat askance at what we term “ungrounded” theoretical work, which we too often dismiss as “philosophizing.” (Philosophy has, of course, its own sets of objects, often the process of thinking itself, but done poorly it does open itself up to having no objects at all.)

Strangely enough, we are more likely to accept work that is at the other end of the axis: folklore studies has a long history of valuing the collection of objects of various kinds. The rationale for such valuation is often twofold: one is the notion of salvage that lies at heart of folklore studies — that the preservation of material that would otherwise be lost to history is an important act, and valuable contribution, in and of itself; the other is that such data is fertile ground for the theoretical development and model-building that will surely follow. Both facets are in fact included in the Journal of American Folklore’s charter published in the very first issue: “it is obviously more important to gather materials which may form the basis of later study than to pursue comparison with insufficient materials; especially as the collection must be accomplished at once, if at all, while the comparison may safely be postponed” (7).

Most work in folklore studies occupies the space between these two poles, with the responsibility falling upon the analyst to decide what matters more to her: the particularity of the data or the universality of the theory. Henry Glassie described himself as an analyst more interested in the former, and it is not uncommon to see folklorists, and other analysts, in fact deriving their theories from the data itself: it is simply a further abstraction from the patterns usually embedded in the data itself. How portable the derived theory is is up to readers to determine, but it is quite common for an idea first articulated in one study to get taken up in another study, and then, through the slow accumulation of citations to develop into its own theoretical nexus.

In fact, quite a few of the bodies of work that we consider to be theoretical in nature really arose because their authors felt that the data before them was either not adequately explained or not addressed at all by the theories available to them. (This might be what the beginning of a paradigm shift looks like in the humanities: a lack of explanation or a lack of coverage. Imagine, for example, being a literary critic in the 1970s interested in Monique Wittig’s Les Guérillères and having only New Criticism available to you think about/through the novel. As a mechanic friend of mine might say: you don’t have the tools for the job. In some cases, some analysts simply wait for the tools to be developed, but other analysts decide to start building things for themselves. Sometimes they continue on their own, and sometimes they are joined by others.

Or sometimes they are part of a collection of like-minded analysts who find that what they are interested in isn’t even conceivable in the current theory (or theories). This is what happened with Richard Bauman, who found himself slowly assembling the pieces of a interpretive and ideational framework that became known as “performance theory” in folklore studies, but it wasn’t long, thanks to the interdisciplinary nature of folklore studies, before it slipped its reigns and became part of conversations in disciplines focused on more traditional kinds of performances, like theater studies, or in more formal kinds of performances, like communication studies. In his observation about the two traditions, Glassie observed that Bauman was an example of someone who enjoyed collecting data but largely saw it as a way to develop, extend, or refine the theory which was his central concern in much of what he did.

And so now you find yourself as apprentice authors in a field like folklore studies, seeking to find a place to start, and more established scholars like your faculty keep giving you what seems like evasive answers which too often seem like elaborate, and occasionally articulate, versions of “it depends.”

Because it does.

It depends on what your own interest and investments are, but you also need to recognize that the axis of attention does demand that any analysis possesses both data and some theoretical orientation. Time is short in a semester, that’s a given, but the press of time sometimes results in people engaging in needless wheel-spinning because they do not have the traction that results from having a clear sense of what their data is or what their theory might be.

You can, however, use this axis of attention as a way to gauge the nature of your project, and perhaps what it requires. If you have only one or two examples of a given phenomena, and that is all you are likely to have, that means your work needs to have a very developed theoretical framework that makes those one or two data points compelling examples of some larger phenomenon. If you have twenty or thirty examples, then it is likely you will require less theoretical orientation and will spend more time in your analysis, compiling and collating materials into interesting categories and trends. (This is still small data by data science measures but fairly large data by humanities standards.)

This also means knowing your own strengths, orientations, investments, interests, and (imagined and/or hoped for) intellectual trajectory — hile we sometimes imagine it as not like those other things, the academy is a kind of marketplace of ideas and approaches, and the work you publish will mark as you as a particular kind of scholar. This is dynamic, of course, and there are plenty of scholars who have changed their research agenda, for a variety of reasons, and enjoyed a switch from one orientation to another. (And I’ve seen it go both ways, so it’s not always towards abstraction.)

Rooth 1980: “Pattern Recognition, Data Reduction, Catchwords and Semantic Problems”

If, like me, you are committed to finding prescient work in the realm of computational approaches to the humanities, it means you are often tracking down somewhat difficult to find volumes and quickly photocopying an article or two while you still have the volume in your hands. Anna Birgitta Rooth’s “Pattern Recognition, Data Reduction, Catchwords and Semantic Problems” is one such article, and the PDF I am making available has been OCRed.

AFS 2018

For those who have asked, below are links to the paper I gave at this year’s meeting of the American Folklore Society along with the slides and the handout (which was a version of the slides, so you don’t need both). As I catch up with everything on which I have fallen behind, I will post my notes about the conference itself in some fashion.

Here are: the paper, the slides, and the handout for “It’s about Time: How Folk Narratives Manage Time in Discourse.”

Abstract: Concluding his consideration of “Time in Folk-Narrative,” Bill Nicolaisen noted that the nature of human experience is centrally of time and that what marked genres of folk narrative, perhaps as much, or more, than anything else, was their management of time: “What must be stressed, however, is that in contrast to the concepts and realization of an extended present and of narrated time in the folktale, the dramatic comparisons made in the legend are designed to demonstrate the incompatibility of the two time frames, which exist as parallel systems” (318). Much of Nicolaisen’s efforts are focused on a careful compilation of how time is signaled, and thus managed, within the discourse of ten fairy tales drawn randomly from Thompson’s One Hundred Favorite Folktales. This paper revisits and extends Nicolaisen’s work, taking as its central task the careful attention to words used. Where Nicolaisen focused principally on the folktale, with occasional references to legend, this paper, part of a larger examination of legends in the current moment, uses a number of legends taken, first, from oral discourse, and then a number of legends found online. It follows this examination with a look at, what the paper itself argues, is the adjacent genre of the personal anecdote, sometimes also known as the personal experience narrative, in order to determine how a close examination of the management of time, in discourse, might reveal where the two genres converge or diverge, in hopes of finding a better way to model both and reliable discursive cues. Some of the methodologies deployed are computational in nature, beginning with forms of markup first explored by computer scientists Pustejovsky et alum and followed up by recent attempts to automate temporal signals in texts by David Elson. The current work seeks to re-imagine the pioneering work of Bill Nicolaisen, and before him Benjamin Colby, in light of recent developments in computational modeling of narrative with an especial focus on what that means for the study of genre.

Nicolaisen, William. 1978. Time in Folk-Narrative. In Folklore Studies in the Twentieth Centuries, 314-319. Ed. Venetia Newall. Rowman and Littlefield. (Available as a PDF.)

The Dissertation as a Book

Much ado over the American History Association’s proposed embargo of dissertations being made available. A lot of of interesting conversations [covered here](, which, perhaps, brushes up against the topic of the nature of the scholarly book. The central question: who are books for? Fellow scholars? Wider audience?

I like Henry Glassie’s stance on this, as he once noted in conversation as I struggled to think the topic for myself early in the process of researching and writing _The Makers of Things_: “Books are for people, articles are for scholars.” You don’t have to agree with him, but you can still admire the precision of his thinking and the realization that there are two audiences. He followed the distinction with the observation that he had never achieved recognition from within the field until he had achieved it outside the field, and so his “breakthrough” books were those that had “broken through” elsewhere, and that *that* was what his (our) colleagues recognized.

This dynamic is more generally true than most of us care to admit. Here at my university, as probably occurs other places as well (but I cannot speak to them), significant raises are only possible through two routes:

1. *sycophancy*, which is the most common one locally, and
2. *portability*, e.g., getting a job offer from elsewhere.

In Glassie’s distinction and in the second route what we see is that internal audiences don’t themselves feel comfortable judging the merit of their own but require an external audience for validation. On the one hand, this is not a bad thing, since third parties might just offer the objectivity we seek, but it is stymying to those who seek a middle way of improving from within according to an external sense of quality. (Confessional aside here: I tried this middle way, and it doesn’t work. At least not in my local environment. Perhaps others have had better luck elsewhere. I get the sense, for example, that a colleague of mine at Indiana University has been very successful in this regard, though he has shown uncommon persistence in pursuing such a path.)

Locally, I encourage my dissertators to write as much as they can as if they were writing a book to be published into the trade nonfiction marketplace. This works in their favor, I think, in two ways: first, it encourages them to think about what it means to write a book and moves them closer to having a manuscript of that nature, and, second, it encourages them to think of the dissertation as a vehicle for communication and not simply display of scholarly mastery. This is important since many, if not all, of my students are working at the margins of traditional literary scholarship, and they need to think about how they are going to position themselves in the marketplace. Sitedness is very important in such a rhetorical moment, and young scholars especially need to be aware of their audience’s expectations, biases, and blindnesses.

The L.A. Times has a short article, with lots of great links, about [the rise in popularity of long-form non-fiction]( If the monograph is dead, as many lament, viva the readable book!

Why Most Published Research Findings Are False

_PLoS Medicine_ has a fascinating article by John P. A. Ioannidis that argues that in an era where all research must establish, almost *a priori*, its “significance,” that we in fact have ended up with research that is insignificant. The problem, as I understand it from my reading, is that too many scientists — and the window onto the scholarly world is open here, I think — are required to be productive in ways that bureaucracies can “measure.” Thus, the race is on *toward* smaller studies that are easily commoditized into publications and *away* from larger studies which either require years to produce results or have too many collaborators for credit to be pieced out in ways that institutions like.

> There is increasing concern that most current published research findings are false. The probability that a research claim is true may depend on study power and bias, the number of other studies on the same question, and, importantly, the ratio of true to no relationships among the relationships probed in each scientific field. In this framework, a research finding is less likely to be true when the studies conducted in a field are smaller; when effect sizes are smaller; when there is a greater number and lesser preselection of tested relationships; where there is greater flexibility in designs, definitions, outcomes, and analytical modes; when there is greater financial and other interest and prejudice; and when more teams are involved in a scientific field in chase of statistical significance. Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true. Moreover, for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias. In this essay, I discuss the implications of these problems for the conduct and interpretation of research.

Here’s the official citation:

Ioannidis JPA (2005) Why Most Published Research Findings Are False. PLoS Med 2(8): e124. doi:10.1371/journal.pmed.0020124

Assessing the Future Landscape of Scholarly Communication

The report should probably really be titled “Assessing the Future Landscape of Scholarly Communication: A Really, Really Long Report” but in fact its subtitle is “An Exploration of Faculty Values and Needs in Seven Disciplines.” One of the disciplines profiled is history, which I chose as being closest to my own field of folklore studies. How long is that one report — one of seven, remember? — 115 pages. Brevity, thy name is not Center for Studies in Higher Education.

[Here’s the index page for the whole report.](

Knowledge for All

The University of Prince Edward Island cancelled their subscription to Web of Science:

This is to inform the UPEI campus community that we have not renewed our subscription to ISI’s Web of Science database (WoS). We realize this is a key research database for many of you and we have taken steps to ensure access to appropriate alternative resources, as well as the WoS back‑files. Late last year we received notification that our subscription price was going to increase by 120%. A number of factors went into the decision not to renew:

‑ a challenging fiscal climate means that we are unlikely to see an increase to Library budgets;
‑ any subscription increase in these challenging times is difficult, but an increase of 120% is simply not acceptable;
‑ we would have been forced to sign a 3‑year agreement, with additional increases in each of the 3 years;
‑ a weaker Canadian dollar would have a significant impact on our subscription costs;
‑ accommodating this level of increase lends credence to the vendors’ business practices and we felt it important to make a statement against these practices (see‑of‑California‑Tries‑Just/65823/ for a recent decision at UC).

UPEI is also leading an effort to create a free and open index to the world’s scholarly literature called “Knowledge For All”. This proposal is currently being sent to various Canadian and international library consortia in an effort to gain support for the project. One goal of Knowledge For All is to ensure that scholars and members of the broader public are no longer disenfranchised by a broken system of scholarly communication. We will provide the campus community with updates on this effort.

It’s interesting to note that it may very well be the smaller universities that make some of these shifts, perhaps clumsily, first because they usually are closer to the economic trends than the majors. I think such is also the case with my own university.

Your Backyard

**Your backyard** is the best place to discover new things, at least in the case of species. At least that’s what Stephen Fry claims in [this episode]( of QI. Apparently a biologist in Leicester performed a long-term study of her own garden and over the course of several years discovered many species that had never been documented in Great Britain and four new species of wasp that had never been documented before at all.