5. The Practice of Research and the Craft of Writing

As the scholar moves further into research, the various stages of a proper method of investigation blend with one another. After a decade of research and writing, one will not spend a lot of time identifying the topic. One’s area of specialization and its sub-­themes become so familiar that the nature and extent of future research projects are easily anticipated. In addition, the work of gathering materials merges with the tasks of writing and outlining. Assuming a broad continuum in one’s general area of research, materials that have been gathered in the course of previous projects are added to a basic research bibliography while, at the same time, the focus of research is changed slightly to adapt to a new project. There is, moreover, no hard and fast rule for drafting an outline or developing a pattern of approach: researchers must think about all aspects of their work continuously, even though they tend to move from a gathering and an evaluating stage to a collating stage and finally to the work of writing and redrafting.

Thus, as we conceptualize a thesis or dissertation, our thoughts naturally range over the entire project, from beginning to end, often (and rightly so) without a clear sense of all of the issues at play or a firm grasp of the ultimate shape of the project. Nevertheless, at the outset of research, once a topic is selected, we must begin by gathering materials and arranging them on the basis of a preliminary outline. Earlier chapters have provided a survey of scholarly tools that enable the researcher to take the very first step of acquiring a working knowledge of the bibliography that pertains to his or her specialization. It may be intimidating, but it is nonetheless true, that research students are responsible for being aware of everything that has been written in their field up to the point of their own approach to the topic. Eventually, they will master the entire bibliography of a field, as represented by the most recent and/or the best writings in their specialization. This mastery, however, is not accomplished by simply multiplying references; the basic bibliographical task always involves a process of sifting and evaluation. We turn in this chapter, therefore, to examine basic techniques and skills in research and writing.1

Evaluating Resources and Materials

Students need to cultivate the ability to identify the important books and articles on a topic and distinguish them from the unimportant ones. This process of identification, moreover, must occur at the level both of primary and of secondary sources. Once the basic bibliographic search is completed, we move toward a process of increasing selectivity. At the level of primary sources, the explosion of documentation brought about by the proliferation of online resources not only raises the issue of narrowing a topic for the sake of finding a legitimate subject that can be addressed at the requisite level of detail, it also raises with considerable intensity the issue of distinguishing materials of greater relevance and significance from those of less importance among the available sources. A researcher must make an effort to identify authors, assess their impact on their contemporaries, and locate their work in its historical context. Not all of the documents produced by presumably important or influential individuals are important documents — and some of the documents produced by presumably insignificant individuals may be important documents. At the level of secondary materials, books that are written at a popular level, or are not documented well, or are not respected in the field, should be identified and set aside. Age can say a lot about the strength of a work, particularly with respect to its bibliographical apparatus. But one cannot simply presume that the older book or dissertation is weak; each must be examined on its own merits. An ill-­documented popular work may actually document a significant moment.

We have already discussed hierarchies of value in the use of reference works, such as encyclopedias and dictionaries. The same general rule applies to journal articles. The best, most recent articles will often add additional bibliographical material that will lead one into the subject in more detail, and thus it is helpful to always work from the most recent articles backward. It is also desirable to start with the most important journals, a datum that is sometimes determined by the breadth of the journal’s coverage. The journal Church History, for example, would commonly possess a higher priority than Methodist History (unless, of course, one’s topic was specifically Methodist history). Articles on Methodism in Church History (if the authors have done their work) will undoubtedly refer to articles in the denominational magazine. Research students should start as broadly as possible with the more authoritative journals and then work their way back, both in time and in detail, as that will save considerable effort along the way.

One should also be alert to the distinction between archival and antiquarian work, on the one hand, and critical scholarship on the other. For example, journals like Notes and Queries contain edited primary texts, but very little scholarly apparatus or critical interaction. In this regard as well, the distinction between primary and secondary sources must always be kept in mind, a distinction that depends a great deal on the intention behind the creation of the evidence and the use to which the evidence is put. We have argued that Williston Walker’s History of the Christian Church is a good example of a secondary, or in fact, even a tertiary source, granting that it does not deal directly with the materials of history and in many places is itself based on what we would call secondary sources. But if one is studying the life and thought of Williston Walker, then it is no longer a secondary source. It is a document that needs to be dealt with differently, in light of the fact that one is going to it, not for the information that it was designed to convey, but for information about the mind of Williston Walker and his various redactors.

Students should also be aware of the distinction between the critical editions of documents that we have come to expect in current scholarship and those found in earlier edited works. Critical editions commonly contain an elaborate editorial apparatus that explains in precise terms the degree of liberty that was taken in collecting, compiling, and editing the documents. Many older edited works, however, especially of diaries, correspondence, speeches, and sermons, leave out entire documents or parts of documents, and sometimes they conflate documents in such a way that the chronology is confused, or even misleadingly altered. We thus distinguish between original sources, whether in manuscript or in print, whether presented in their original form or on film, fiche, or electronic format, that remain unedited, carefully edited critical editions that reveal their editorial process, and more or less uncritical editions in which undocumented alterations (whether textual or typographical) have been made. We similarly distinguish between denominational and antiquarian journals that often contain primary sources and major journals that seldom do, and we use these sources differently. Research students will also gradually develop a sense of which journals are refereed by specialists and this knowledge will help place a journal in an implicit hierarchy of authority. Some journals will almost always accept an article, while others require the review of two or even three scholars in the field, and the latter commonly reject more articles than they accept. In time, a person will gradually acquire a sense of which are the standard scholarly journals and which articles are definitive in their field.

Taking Notes

The specific method of note taking will inevitably be tailored to the individual and to the precise requirements of one’s own research project, but certain general guidelines can be laid down. The cardinal rule is to always record notes in exactly the same way and to follow an approach that performs the basic task of identifying sources and resources correctly and in a usable manner. One may use discrete and well-­identified files on a computer or an older, hard-­copy method such as blank sheets of paper cut to the size of file cards or 6 × 9½ narrow-­lined loose-­leaf paper which goes into standard one-­inch and two-­inch notebooks.

Whatever the mechanics of one’s research, the important thing is to be consistent and to record information with exceeding care. With any system that is adopted, and with any medium that is utilized, the materials that one chooses must be designed so as to fit into an overall filing system that can store several projects in good order, suitable to the delineation of discrete units of text and amenable to the conflation of the units of text into larger sections. In the older, hard-­copy model, consistent use of sheets of paper that are all one size allows notes taken for one project to be conflated with notes for another. For example, a sheet taken out of one’s church history lectures should ideally be able to fit directly into research on a book. The same principle applies even with the use of lap-­top computers and databases: the consistent use of a standardized pattern of naming and organizing files is absolutely necessary. Files need to be named in such a way as to be capable of retrieval and use — i.e., one needs to know where, in an increasingly vast memory space, one’s files on a particular subject are kept. And one’s computer folders need to be as carefully organized as hard copy folders in file drawers. Each distinct project ought to have its own distinctively named folder and each section or chapter of a project ought to have its distinct name, related to the topic of the project. If, for example, one were writing a book on Luther’s understanding of marriage and divorce, there ought to be a main Luther folder and there ought to be files within the folder named accordingly, for example, “Luther-­marriage01” and “Luther-­marriage02.” A file named “marriage01” could all too easily be copied by accident into another folder, say, dealing with Bucer’s views on marriage, and copied over a Bucer file named “marriage01.”

In the old manual method, one would never put more than a paragraph or a single idea on one sheet and would never work on the back of pages. On a computer, one ought to maintain discrete files for discrete subjects: notes taken on one volume ought not to be included in the same file as notes taken on another volume. In the case of the more traditional method of note taking, the reason for using standard paper or paper cut to size and not the commercial 5 × 8 index cards is that with the latter the bulk of paper soon becomes too great. Index cards are probably three times as thick as twenty weight plain bond. It is also a great deal less expensive to use standard blank or lined paper, and there is an important psychological rationale to this technique as well.

If notes are to be collated easily, there must be no more than one basic idea or paragraph on each piece of paper or, in the case of files on a computer, the researcher should attend to careful paragraphing within a file. Those who use index cards, however, will be inclined not to waste them. Scholars should be willing to waste paper, and if they must err, they should err on the side of using an excess of paper, because it will enable them to avoid the time-­consuming task of using scissors and paste and cutting up the note card. Ideally, paragraphs written one year should be able to be inserted between paragraphs of materials that were recorded in other years, or at another stage of research. This is also the reason for using paper that is no larger than 5 × 8 or 6 × 9½ and for not working on the back of notes. This logic carries over directly into research and note taking on a computer. There is, of course, no issue of physical bulk, but there is the issue of discrete bits of information that can be maintained and managed as discrete bits, identified easily by the name of the file in which they have been saved, and then retrieved for the sake of collation with other discrete bits of data. A large number of ill-­named and badly organized files is the electronic equivalent of a mass of unorganized sheets of paper. When files are well-­organized and careful paragraphing done at an initial stage, the “scissors-­and-­paste” approach to collating is easily adapted to the block moving of text from a research file into an outline.

Notes must also be properly documented. Some scholars utilize a footnote apparatus on each page. Others use a slug with the author’s last name and short title in the upper left hand corner. All scholars, however, use separate file cards, usually 3 × 5, for complete bibliographical information — author, title, place of publication, publisher, date — and file them separately. In a computer model, each quotation or paraphrase taken down as notes on a single source ought to be footnoted precisely, as one would do in the final draft, including a full reference in footnote-­form in the first note taken and standardized short references in subsequent notes. Do not use “ibid.” for short references. Always use a standardized short reference, preferably containing the author’s last name and a first key word or shortened form of a title. This rule is of particular importance when taking notes or writing on a computer, inasmuch as footnotes travel with the block of moved text and the source of an “ibid.” detached from its original context may be very difficult to identify.

Cultivating the habit of taking bibliographical information down in proper bibliographical form is essential, with the last name first, then the first name in as full a form as possible, the title, and so on. A strict regime developed in the initial stages of research will serve one well when the final stages of the bibliography and dissertation or thesis are reached. The very first thing to record on the individual sheet of note paper is the author and a short title of the book or article that can take one back to the full reference — just as the first thing to be recorded in research done on a computer is the full bibliographical reference. If this is conscientiously done on each initial sheet or at the beginning of each discrete file, the painful and discouraging task of being forced to redo one’s work can be avoided. In addition, as first references are taken down, a bibliographical file ought to be started in which the primary and secondary sources for the project are listed alphabetically in full references. Given that most projects begin with proposals, the researcher will most probably have a preliminary bibliography at the outset of research that can simply be augmented as the research progresses and new resources are identified.

When working with Internet databases, bibliography should be up-­ or off-­loaded in a text format that can be edited and moved into one’s bibliography. In the case of drawing bibliography from the online catalogs of major research libraries, the “full text” or long-­form reference should be taken, not the short reference, given the considerable amount of useful information found in the former. A long-­form reference will often include secondary authors, indications of problems in a text, and comments concerning variant editions, anonyms, pseudonyms, and other significant information compiled by librarians, archivists, and catalogers. The long-­form reference can be kept in a bibliographical file and the short reference extracted for one’s own bibliography.

The distinction between an accurate quotation from a work, a paraphrase of a work, and one’s own analysis is also crucial. At the very outset of an encounter with a document, the question should be put, “Is this worthy to be quoted in full or not?” The usual answer should be “not,” in that the number of absolutely memorable quotations is few. Then as one studies the document and begins to take notes, it is useful to ask, “Am I best served with a paraphrase, or analysis, or both?” and “To what extent am I borrowing the phraseology of the source?” Notes must make the answer to these questions very clear, because a year from the time the note is recorded, it may be impossible to recall whether the words are the student’s own, or whether they are those of the source.

Generally, fewer notes are taken on secondary sources than on primary ones since the principal arguments and main conclusions of secondary sources, such as articles in the field, should already be very familiar. In the case of scholarly articles, which can easily be downloaded in “pdf” or “gif” format or simply xeroxed when the electronic form is not available, the researcher can literally own the materials and consciously reduce the bulk of notes. In the case of primary sources, even though these too can often be downloaded, note taking on a larger scale is necessary: notes on the primary sources will become the core of one’s final essay. While taking notes, whether from primary or secondary sources, discipline is required to avoid a pattern of passivity. It is a great temptation to put off creative writing until a segment of research is completed, but the temptation should be resisted by requiring oneself to record any insights that have arisen in the course of research. These insights often become the nucleus of a thesis, or part of an introduction or a conclusion, when the project matures to the point of serious writing.

In addition to the temptation to be passive in research, there is also a tendency to overquote material. Beginning students often feel an internal pressure to get the whole text in hand and get it right, but they should steel themselves to put the matter in their own words, being careful, as already noted, to record what level of generalization they have already adopted. Inevitably, the day will come when a needed quotation is missing, and the only alternative will be to go back to a document and pull something out. But slovenly habits that necessitate such retracing of one’s work should be noted and avoided. Part of the skill of research involves developing the ability to interact analytically with a document, rather than simply reproduce it. The task of the historian is not to reproduce some old document; that is the work of the antiquarian; rather, it is a matter of discovering what is really important in the evidence and eliciting a few supporting phrases or quotations. Decisions need constantly to be made concerning the nature of one’s representation of a source: is a quotation necessary; ought one to paraphrase, summarize, or allude — with, of course, a proper footnote identifying the source. If there is a document-­specific or period-­specific or topic-­specific technical vocabulary in the document, one needs to understand it and reflect that understanding in the way one handles it. In some cases it may be that the vocabulary in the document will need to be reproduced very closely. But by and large, the main endeavor will involve analyzing the evidence and critically dealing with the content of the document. Even at this first stage of research, students should move toward an informed dialogue with the primary evidence and the secondary authorities.

Collating Notes: the Preliminary Outline

Besides attending to insights that arise in the very process of research, one should constantly be giving thought to how the material will be ordered. While engaging in basic research, reading the so-­called “primary” and “secondary” materials, historians and theologians should be thinking all the time, “How does this fit together?” The preliminary outline both for individual chapters and the entire dissertation arises from at least three places. First, the initial consideration of the boundaries of the topic offers some guidance. Second, the outline emerges from the materials themselves and what they provide concerning their chronological or logical flow. Third (and less importantly), the outline arises from secondary sources and other people’s opinions about the materials. The last is the least satisfactory place to obtain an outline, although it is going to have some value and influence, particularly since we are entering a discussion about the material that has been going on for some time.

In the beginning stages of research, it is highly desirable to make a clear decision about the best kind of outline to adopt and its suitability to the materials. Is it a chronological/historical outline? Is it a logical/topical outline? Or does it combine patterns? One may opt for a primarily logical model and then, within the topic and its subsections, proceed chronologically. Or one may chose a logical model and simply move systematically through the topics. On the other hand, a predominantly chronological model may well be best, with logical sections in it. The outline may also be arranged in a third pattern: it may arise from the argument that has been joined with previous scholars. If one chooses the third pattern, then research becomes pieces of evidence for and against a case. In most scholarly work, at some point, these three forms are combined or converge. The one question that must be asked at the very beginning is, “How is this material best organized?” In some fields, such as the history of philosophy or intellectual history, a purely chronological pattern does not often work, granting that (to disagree with some famous general) history is not just “one damn thing after another.” A simple attempt to move from a document written in 1595 to a document written in 1596 to a document written in 1597 will almost certainly fail. The important line of investigation may lead from the document written in 1597 and then to its antecedent in 1596, along with other documents written in close proximity, and finally to a heretofore unappreciated author who wrote several years later. In intellectual history and theology, chronology is not necessarily the best way to proceed, although one must always be sensitive to it, granting that influences normally happen in roughly chronological order.

The first outline is necessarily preliminary because in the process of working on additional documents, further secondary materials, and even writing the essay, one’s sense of the shape of the project is going to change, sometimes more and sometimes less substantially. The outline is a kind of skeleton statement of a working hypothesis. The hypothesis is about how the materials fit together, and it can, and indeed should, be changed even up to the very end of the project. As a method for structuring complex material, the outline is absolutely essential, but the importance of one particular form of the outline should not be overestimated. A carefully thought through preliminary outline, however, must be in place at the beginning of one’s work because the outline gives shape and direction to one’s entire effort. It will also provide the basis for one’s discussion with his or her primary mentor.

As the outline develops and expands, one is then in a position to organize notes around the outline as the research proceeds. Every note card should reserve the upper right hand corner for a slug that indicates the part of the outline into which the material fits. In the very first stages of research this may be little more than an informed guess, and one should not expect too much certainty about the placement of some bits of material until the research has progressed considerable distance. But the process should be started as soon as possible. When the outline reaches a more complete form, one should go back through all of the research files or, if the research was done manually, the file cards, entering on them their location in the outline with Roman and Arabic numerals. This process will be repeated for each chapter. The last step involves taking the whole roster of files or the stack of note cards and organizing them according to the outline. Computer files can easily be renamed or, if original filenames are maintained for the sake of overall clarity in the project, prefixed with a numerical system to determine their new order.

One is then in a position to begin writing. The steps in this process are essential, since it is really the only way to control vast amounts of material. There are, however, several possible variations to the pattern. There is a traditional manual method that worked well in the days before computers that can easily be adapted to use on a computer. The main divisions of the outline can be expanded onto individual sheets. With the heading of the outline on these sheets without anything else on the paper, the materials of the chapter or section can be collated behind those sheets. The sheets become the place where one writes down an introductory paragraph to that particular section, or a transition from the previous section. A word processor makes the process simple and elegant: a copy of the entire outline of the project can be coded as headings for a table of contents, so that generation of the table will produce an identical outline. Then the materials, primary and secondary, can be copied from the original research files into their proper places in the outline. (The original research files should be left intact for future reference.) The coded outline becomes the shell of the whole project. Pieces of research can be moved at will, including whole sections and sub-­sections of the project, together with their coded headers. The table of contents feature of the word processing program can then be used to generate a new table of contents that includes revisions of the section and sub-­section titles and rearrangement of sections and sub-­sections. In such a process, it is wise to retain not only daily backup files of the entire project as it is developed and expanded, but also to retain old backups, both of the original form of the outline and of earlier forms of sections and sub-­sections — just in case one regrets a change and needs to reconstruct an earlier version or shape of the project, whether in whole or in part.

When writing chapters that extend to forty or fifty pages in length, one simply has to have a method for connecting the complex materials of research coherently to the overall project of the dissertation. The outline not only serves this purpose, but as continually revised to account for new insights into the meaning and interrelationships of materials, it will aid in the process of developing a coherent argument. The skillful use of a well-­conceptualized outline is the only convincing way to relate a thesis to its supporting data. In fact, if the outline has been conceived as a skeleton or as a summary form of the thesis and argument, its alterations and refinements will consciously reflect alterations and refinements of the thesis and argument of the project.

Survey of Secondary Literature and Paradigms: Writing the First Paragraph or Chapter

The formal and conceptual difficulties encountered in framing and writing a good first paragraph to an essay or first chapter to a dissertation or book are substantially the same. In the latter case the first chapter is just developed on a larger scale. In both cases, the fundamental task of the writer is to offer a preliminary conception of the whole project framed by a clear summary-­analysis of the present state of the scholarly question (often constructed as a history of scholarship), a concise statement of the thesis or argument to be presented in view of the history of scholarship and the researcher’s own insights into the primary sources, a statement of the specific way in which the project proposes to advance the scholarship and alter the state of the question, and, optimally, an indication of the method or approach to be taken in the project. In the case of a lengthy and involved project, an analytical statement of the shape of the project or an analytical prospectus of the chapters may be desirable. Of course, the project may change — it probably will change — in the course of writing. Still, a preliminary draft of a first paragraph or chapter is as essential to framing a project as the careful rewriting of that paragraph or chapter will be to its completion.

In the introduction of any topic, questions of secondary literature and paradigms arise, and they bear directly on how the first paragraph should be written. The most compelling way of writing a first paragraph or framing a first chapter is to commence the argument by engaging with other writers in the field. Once the outline is in place and the note cards are collated, a person should have a good grasp of the larger meaning of the topic and its overall shape. By this time one will also have a preliminary sense of the way the thesis fills a gap in the literature and the precise nature and location of the disagreements that have arisen in the course of research with what other people have said about the topic. These considerations yield up the material for a very sound first paragraph of a scholarly article, or the entire first chapter of a dissertation. The first paragraph or chapter should lay out the secondary literature in an analytical form. The case, of course, will be argued on the basis of one’s own first-­hand acquaintance with primary source materials, but while these materials are implicitly present in the opening paragraphs, they are not displayed in detail at this point.

The history of scholarship should be ordered paradigmatically. At a certain stage in research, leading towards the first paragraph or chapter, one should be able to say that certain lines of scholarly argument can be identified and separated. Certain authors fall into place with one line of argumentation or method and other authors with another. The first paragraph or chapter will show that there are several different ways of looking at this material. Having established that there are numerous different treatments of the material in the past, one proceeds to say how these different approaches disagree and which ones have particular insights, and which ones miss issues. After the entire paradigm of previous research is surveyed, one can say, “Granting that scholars have seen this group of issues but have missed these other questions individually, we now have something further to say about this, which moves us off in another direction covering the areas that have been missed.” The thesis of the article or dissertation thereby arises directly from critical engagement with the history of scholarship and the knowledge of new or differently construed primary evidence.

Constructed in this fashion, the first paragraph accomplishes a whole series of different important effects. Effect number one is that it crystallizes in the student’s mind their own purpose for writing and the distinctive character of their own approach. Previously, it is entirely possible that one may not have fully recognized the value of their own research and its place in the history of scholarship, since to that point they had not articulated it in detail. Secondly, in terms of persuasiveness, this paragraph or chapter proves to the reader that the author knows the material. When submitting a dissertation to a committee, or an article to a journal, or a book to a publisher, beginning scholars face the difficult task of convincing critical readers that what they are saying needs to be said. Those readers, if well chosen, will know the literature and will know the field. The longer it takes one to convince the referee, the less likely it is that the essay will be accepted. If, however, readers are convinced in the first paragraphs of an article that the author understands the literature in the field, that he or she has seen a gap in it, and that there is something important that can be said to fill the gap, then one can be assured of the acceptance of an essay or article. If one expertly surveys the literature in the first sections of the first chapter of a dissertation, one can most probably rest assured that both internal and external readers will approve it. The paradigm of past scholarship demonstrates that the student has not only exhaustively covered previous research in a specific area, but the compelling way that this material is ordered suggests that the author also possesses a commanding knowledge of the field. Scholarly authority is firmly established, both on the grounds of comprehending an area of research and on the basis of the intellectual power that orders the material.

Once the first paragraph or paragraphs are completed, one not only has a good introduction to the essay or dissertation, but also the beginning of a sound methodology for writing — namely, rigorous interaction with secondary materials in the text of the essay. The body of the essay or chapter should recognize the secondary literature specifically; one should not have to go to the footnotes to locate dialogue with other scholars. This discussion with other authorities arises directly from the introductory paragraph where the issues are set up initially, and this, once again, demonstrates a student’s competence throughout the essay. But it does something even more than that. It proves a scholar’s own grasp of the material by forcing us to enter into conversation with other people who have written about the topic. This needs to be consciously done. The initial survey of scholarship introduces the topic and the issues surrounding it, but the discussion with other authorities is something that must be carried on throughout the entire project. Otherwise a student may end up writing a monologue that does not necessarily add anything to the history of scholarship.

Some final advice on the first paragraphs pertains to the initial efforts of written expression. Authors commonly agree that the opening paragraph or two is the most difficult part of any project to write (for each chapter or for each article). It is conceptually the most difficult part of writing because the whole project must be held in the mind at once, and at the same time, the discrete parts of the work must be described and an effort made to order them in a compelling way. One may have completed their research and come to write the first paragraph and be seized with a tremendous case of writer’s block. Recognizing the conceptual difficulty of constructing the first paragraph may help, and it is also an aid to composition to gather together in one place the insights written out during the process of research and see how they may bear upon the introductory sentences. If progress is seriously stymied for any length of time, it may be best to pass over the first paragraph and begin writing out the evidence that supports the thesis. Once the various parts of the argument are written, one will often return to the first paragraph with greater clarity.

As one begins to write the first chapter, research notes are folded into the discussion with the secondary sources. We begin to draw together in one place the discrete units that become the basis of paragraphs. Most of the time, the order of writing will follow the subsections of the provisional outline. It may, of course, be desirable to rearrange the outline a bit after the introduction is finished. Then we proceed to write through the outline, usually starting with the first sub-­topic. As one progresses, the bits of information on note cards or in files on one’s computer (or in whatever medium employed) are rewritten and put into decent English. By this point one should have a sense of how these single pieces of research can best be fit together, either through the logical or historical connections between them, or by means of the arguments that are joined with other scholars.

Students who have undertaken advanced research are sometimes overwhelmed by the thought, “How do I write a book?” The answer is, “You do not.” Nobody writes a book. All scholars do is write little pieces of a book, and they keep on writing little pieces with some sense of the whole given by the outline. Finally, after writing a whole series of small sections, they assemble them into a cohesive whole and make a book. The object is to take a discrete portion of research or writing at a time, engage it rigorously in the light of one’s sense of the whole, then move off of that topic and do more. If a student writes one subsection every week or every two weeks, then in a year or so the result is a book-­length manuscript. The task involves moving through the individual elements, never losing sight of the larger whole, and hopefully allowing those smaller units of information to shape and alter the project in the process.

The Body of the Essay

As they progress through the various chapters, research students should give conscious attention, not only to the substance, but to the tone of their work and what they are attempting to prove and why. Their work should be primarily constructive, rather than destructive, though in most fields there will undoubtedly be some underbrush that needs to be cleared away. Attacking previous authorities, however, is often needless, and as one experienced teacher put it, such attacks are “both irritating and boring to read,” to say nothing of being a waste of space.2 Undue criticism of past scholarship, besides being ungracious and distasteful, is a certain sign of vanity, a quality that young scholars in particular should seek to avoid.

The body of the essay ought to be constructed with clear reference to the developing outline of the study. First and second level subheadings from the outline can and ought to serve as explicit markers throughout the essay. Subheads are invariably preferable, for both brevity and clarity, to transitional sentences or paragraphs inserted only for the sake of indicating transitions to the reader. (In a word-­processed document, all such headings ought to be coded for a table of contents and the table of contents generated anew as a revised outline whenever sections of the essay are rearranged.)

Maintenance and use of a clear and suitable outline will also be of service in the establishment of a narrative or line of argument in one’s essay. The logical or chronological flow of materials adopted in the process of organizing and outlining the materials of research ought to provide a foundation for argumentation — and argumentation should flow smoothly between the introductory paragraphs and the conclusion. Attention not only to style but to the larger structure and flow of argument is indispensable: each section or subsection of the essay should read as a focused unit with its own proper beginning, middle and end, and all of the sections should be arranged in such a way as to facilitate the argument and to draw the reader in a convincing manner through the body of the essay toward the conclusion.

There are several ways to achieve a fairly unified and coherent narrative line. A historical essay can observe chronology and follow out the historical line of a series of events or documents. Care must be used in this approach, however, inasmuch as mere chronological precedence does not indicate the kind of relationship that may exist among documents or even that there is a relationship. Thus, a researcher must consider such questions as whether the writer of one document actually knew of an earlier document, and if he or she knew of the earlier document, how or if that knowledge actually impinged on the writing of the latter document. Another approach, of a more topical nature, useful for the creation of coherent subsections, involves the use of a particular document as the basis for a narrative line. This approach, too, must be used with care. The researcher must ask whether the document itself, in its contents and organization, supplies all of the key issues needed to interpret it: must other elements for the narrative line be drawn into play from the historical, cultural, or social context of the document and of its writer? Are there issues that were self-­evident to the writer of the document that are not self-­evident to the present-­day reader? In sum, the historian must step past the basic question of the meaning of a text to the question of how a text, texts, or other evidences ought to be presented to the end of constructing a line of narrative or argument that is both interpretive and coherent.

A word of caution and advice is also due to those who have never made the transition between research and writing by hand in pen or pencil and research and writing on a computer. Although the computer facilitates moving and editing text in a marvelous way, it also brings certain dangers and drawbacks. The traditional manual methods enabled (or perhaps forced) the mind to work through materials with great care at several different stages in a project. The entire rewriting or retyping of a page, although tedious, typically served as a moment in which the materials were reexamined, insights gained, composition smoothed, and the project advanced in subtle ways. With a computer, however, entire pages are virtually never redone, certainly not for the purpose of minor rearrangements or minor typographical corrections. The researcher must take care that the ease of the editing process does not short-­circuit the necessary work of rethinking a project on the way to second and third drafts.

Writing the Last Paragraph

When one has finished writing the body of a chapter or article, the last paragraph should generally understate the importance of one’s discoveries. Modesty in conclusions is the best rule. At the end of a long, costly research project, students are characteristically enthused about their accomplishments; they may be tempted to think that they have determined the field for ages to come or broken significant new ground. Such enthusiasm commonly leads one to conclude more than he or she has actually proven. On the first attempt, therefore, one should exercise restraint and put the question, “What are the minimal conclusions that I can draw from this?” Understatement is an art, but it is an art well worth cultivating.

Secondly, while framing the last paragraph, ask specifically, “Exactly what did I argue in the essay?” Never draw conclusions for a larger field than the field of study that has actually been surveyed in the chapter. If the study was on “The Doctrine of God in Reformed Protestantism 1570-1640,” do not conclude anything about the doctrine of God from the early Reformation. Do not conclude anything about the doctrine of God for later Protestantism or about Lutheranism. Do not conclude anything about the whole system of theology from arguments about the doctrine of God, except insofar as that doctrine obviously relates to it. Or if one is studying church polity in Geneva at the time of Calvin, do not draw conclusions for Presbyterian polity in general. The conclusions should fit exactly the bounds of the topic examined; the research itself thus sets the boundaries for the conclusion. The generalizations made in the conclusion are the generalizations that apply directly to what was attempted. Otherwise students are liable to be wrong — very wrong, possibly even absurd. It greatly devalues research if beginning scholars do not deal expertly with their conclusions, and readers will be quick to notice this failure of discipline. Seldom, if ever, will modesty be criticized.

Conclusions may state implications for further research, but great caution should be exercised in this regard as well. It is best to deal with the possible implications for further research in an identifiable subsection, or simply hold these materials for future projects, since the motive for stating the possibility of further implications is often indistinguishable from the motive that leads to sweeping generalizations. This advice is directed primarily to young and beginning scholars. As one matures, judgment matures, and when a person’s general command of a field becomes highly competent, then more may be claimed. This is one of a number of things that change over time. For example, beginning students often tend towards monocausal explanations of historical events. As their judgment matures, they begin to see the complexities of historical causation, and this leads them to become more tentative. Time spent in an area of research thus leads to maturity which in turn produces more modesty and tentativeness than is often true of beginners. Similarly, one’s command of the bibliography in a field will grow with time and this will broaden a student’s preliminary survey of scholarship and their initial sense of the issues to be addressed and what has been said previously and what has been missed.

Conclusions, in short, should be “analytical” or “analytically descriptive.” The student moves beyond the evidence to make an attempt at saying what the evidence means in a particular context. One can look at an individual writer or a group of writers on a topic, and then say what the structure and the various sources reveal about the breadth, the limitation, and the implications of their thought — where it came from, what it was doing, where it was going. But one should never say that this is the only way to describe the area, nor imply that the field is now exhausted, in a kind of prescriptive definition of what must be noted about the phenomenon. Rather, one moves off of the description analytically, in terms of what can indeed be said about the shape of the phenomenon. The conclusion does remain somewhat open-­ended, although one should have achieved a level of analysis that has not been attained before. The goal is to move off of the material, throw new light on the evidence by means of a different set of antecedents and new material, and then say what this means for the phenomenon in brief, rather than make a strictly prescriptive statement about what must have happened.

Footnoting

The purpose of footnoting is to enable a reader to retrace accurately, if not always in every detail, the steps of a scholar’s research; the reader should be able readily to find the evidence upon which the argument is based. Most research students will be familiar with this principle. It is less likely that beginning students will have given any thought to the structure of footnotes. In writing essays, articles, and chapters there is a fairly predictable structure of footnotes that is related to the way that we set up the initial paragraphs. The initial footnotes, to borrow a military metaphor, will provide a basic deployment of forces to be used in argument while the following sequence of notes that runs throughout the essay ought to evidence a tactical use of those forces, both primary and secondary.

An article will therefore be very heavily documented in the first few pages because we are dealing with an entire field of literature. It is not uncommon for the first footnotes to take up more space than the text of the first two paragraphs. In surveying the field, one is dealing mainly with secondary sources. Once the current state of research is laid out and the thesis stated, the discussion moves on to primary sources. Generally speaking, in the ordering of materials, one moves from secondary sources into primary sources. As one begins to deal with primary documents, the chapter may once again require heavy documentation because of the initial citation of these sources, or perhaps the need to mention technical matters of method. Having examined the primary sources, the argument commonly turns to the substantive results of research, and typically, one will be documenting less and less along the way, because more attention is given to analysis. By the time the conclusion is reached, the documentation will dwindle to the vanishing point, and one is then able to state the conclusions on the basis of the material examined. In most well-­written articles and chapters, one can discern this basic, predictable structure of footnoting and documentation. At the beginning of their research, students would do well to carefully examine the structure of a scholarly article, or a chapter in a well-­received book, and consciously emulate these models.

It may happen that following the introductory paragraphs, and after surveying the scholarship with an extensive apparatus of secondary documentation at the beginning, one discovers, toward the middle of the chapter, that they have accumulated another large body of secondary documentation. In this case, the student or reader should ask whether his essay is in fact two distinct pieces, each with a separate apparatus. If it is one piece, has it been badly organized? Why does a new line of scholarship suddenly appear? Why did I not discuss these matters in my introductory paragraph and fold these references into the initial survey of scholarship? If the answer is that there are two basic issues addressed in sequence in the essay and the form of the essay is clearly reflected by the division of the bibliography, or if there is some other logic that explains the anomaly, then the problem is solved. Such structural oddities, however, are often an indication that the student has inadvertently omitted something from the introductory discussion and inserted it at a later point, or that a second, unrelated topic that really does not belong in the chapter may have been added. Questions such as these should be asked throughout the process of constructing an essay.

The standard work on research by Barzun and Graff may justly be criticized on the question of what goes into a footnote. Barzun and Graff disagree with the maxim that if one wishes to elaborate a point in a footnote, it is either worth saying in the text (and not in the footnote) or it is not worth saying at all.3 They cite several examples of where it is useful to elaborate a theme in a footnote, but the very nature of the exception, and the kind of exception that has to be shown, demonstrates the validity of the rule. In most cases it is inadvisable to develop long excurses in footnotes; if they are really important, they belong in the argument in the text of the paper. Less experienced scholars in particular tend to analyze primary sources in the text of their essays and to dialogue with secondary sources in the footnotes. This approach leads to long footnotes about the patterns and approaches of scholarship — and to a corresponding lack of dialogue with scholarly opinion in the context of the analysis of sources. Observations about the tendencies of scholarship, however, if they are really substantive, frequently belong in the text of an essay as part of the basic analysis of sources, with the footnote indicating where the secondary reference may be found. Kitson Clark was right to suggest that a long, argumentative footnote is often a symptom of mismanagement, and that part of the argument has been displaced from where it naturally belongs.4

Even so, footnotes should reflect the flow of the argument and normally not add new pieces to the essay’s line of reasoning. It is just as important to engage other scholars in the argument as it is to present primary sources and analyze them: the line of reasoning in a scholarly essay, thus, extends the line of the history of scholarship. Foreign language quotations are an exception to the rule and do belong in footnotes, especially when a new translation of an original source appears in the text. Lengthy, untranslated passages may be a sign of erudition, but they also tend to disturb the line of argument, particularly granting that the translation itself is part of the interpretive process of the essay. Citation of the original text in the note offers readers the opportunity to check both the translation and the argument of the essay. A similar exception is properly made when a sentence or two is required to explain a necessary technical point, or to defend the choice of one edition over another acceptable edition of a primary source: these issues belong in a footnote.

This advice cuts against the characteristic temptation of the beginner. Graduate students wish to show that they are comprehensive, and while this is a good basic instinct, the discipline of excision needs to be exercised. One of the hardest things for a young writer to do is to blot out what needs to be blotted out, and to reduce what needs to be reduced. If it is not important enough for the text, students are well advised to transcend their instinct to preserve their work in a footnote. The fact that the student is acquiring information that is new to him or her, but that is by no means new to specialists in the field, means that invariably a large body of notes, and possibly even long, well-­crafted sections of a chapter, will have to be left out of the final draft. Certain elements of background material can, of course, be utilized, but the distinction between what is new to the student and what is new to the field must constantly be kept in mind.

The question of whether or not to footnote material that is used for general or background information is not easily answered. Commonly, we footnote discrete bits of information, but if the task is focused on one particular source in detail, then the footnote can come at the end of the paragraph, because the whole paragraph identifies a unit of information. If, however, the paragraph is a compilation from an analysis of a whole series of different documents, then each discrete bit should be footnoted. Matters of general information, or matters of common knowledge, will often be covered in an introductory paragraph where the field and what has already been written about it are identified. If background material is extended over several pages, it is probably best to footnote it by saying, “The general content of this and the following three paragraphs are drawn from” — and then cite several of the best, most recent secondary sources on the topic. This form is to be preferred to citing those sources over and over again with a string of “ibids.”

As a general rule, all sentences that include quotations from sources, either primary or secondary, ought to be footnoted. This practice prevents a confusion of sources or a question about the identity of sources from arising. When a document is being paraphrased, a useful practice is to add a footnote when the paraphrase passes from one page of the document to the next, but to also offer at least one footnote per paragraph. A paragraph without footnotes, to borrow a rabbinic maxim about the vowel points of the Hebrew Bible, is like a horse without a bridle.

When citing primary sources, it is important to indicate whether the source is an original printing, a standard edition, or a translation, with priority being given to the original printing or standard edition. Thus, an initial reference to Calvin’s Institutes should first cite either the Latin text in the Corpus Reformatorum or in the Opera selecta and then indicate which of the standard translations, if any, has been followed or may be used by the reader. If the translation has been followed, the author should indicate this and indicate, as well, if he or she has made emendations of the translation based on study of the original. Subsequent references may simply cite the Institutes by book, chapter, and subsection. In the case of citations from Calvin’s commentaries, in which there is no standard system of citation apart from the pagination of various editions, the first reference should indicate both the edition cited (again, probably, the Corpus Reformatorum) and the translation used. Subsequent references should include, first, citation of the relevant volume and column in the Corpus Reformatorum and, if the translation has been followed or emended, a reference to the volume and page of the translation, as a courtesy to the reader. Scholarly essays typically should not cite only the translation: this is acceptable only if the point being made does not depend on a close exegesis of text and, in addition, if the essay is not a closely argued commentary on the particular source in question.

Footnotes ought also to be constructed with careful consideration of the method and logic of the work being cited. The practice of citing Calvin’s Institutes by book, chapter, and section, for example, is not unique to Calvin but is the standard practice for the citation of classic texts organized into such patterns and available in many editions and translations. It would hardly make sense to have a footnote that read, “The Bible (Chicago, IL: University of Chicago Press, 1935), p. 510,” when the purpose of the reference is to direct the reader to Psalm 23:3 in the so-­called “American Translation” of Smith and Goodspeed. Book, chapter, and verse must be used in citations of Scripture, granting the vast number of texts and translations. The principle also applies, however, to such works as Calvin’s Institutes, Augustine’s City of God, Irenaeus’s Against Heresies, and so forth — although, in the case of the latter two works, each reference ought also, as a matter of courtesy, to include a citation of the volume and page or column of the edition or translation being used.

By way of summary, there are at least four distinct functions of footnotes, each of which will be found in a well-­organized and argued essay. First, the footnote apparatus will reference the history of scholarship or “state of the question” section of the essay and serve as the basic deployment of primary and secondary sources. These initial footnotes provide the full references to all or virtually all of the materials of the essay, indicating previous scholarship on the subject of the essay, significant studies of the background of the essay, and the primary sources to be examined, whether archival, original printed texts, or critical editions. Second, footnotes function to identify the sources or grounds of the argument of the essay, namely, short form references to the specific locations of cited, paraphrased, or analyzed texts within primary sources. Third, they serve to identify locations in secondary sources that are referenced in argument, pro and con, or as confirmations of a point: these secondary references function either in relation to the argument of the text or body of the essay or in relation to primary documentation concerning the argument and are cited in the same note together with the primary source. Fourth, footnotes serve to identify materials either primary or secondary that represent and offer either amplificatory materials not discussed in the body of the essay or collateral lines of argument not drawn out in the essay. Note that this last use stands as a sound alternative to the use of footnotes for excurses.

Finally, in all footnoting, there is an aesthetics of form and consistency that ought to be observed. Old forms of reference like “op. cit.” and “ibid.” ought to be avoided, not only in the final form of one’s text, but also in the process of taking and collating one’s notes. These forms are particularly dangerous in an age of word processing, in which sentences and paragraphs can be moved at will, including coded references: a stray “ibid.” removed from its original location and proper antecedents can cause much difficulty! It is advisable and a courtesy to the reader to use standardized short form references for repeated sources, e.g., “Calvin, Institutes, . . .” or “Luther, De servo arbitrio, . . . .”

It has become standard form according to many manuals to omit “p.” or “pp.” when referencing pages. Notably, the major periodical database, JSTOR, has rejected this particular alteration of traditional format and continues to note “pp.” for pagination. There is, moreover, a good reason to follow the traditional pattern: omission of “p.” and “pp.” complicates matters when citing medieval or early modern texts, given that some are paginated, some are foliated, and others printed in numbered columns. At least in the cases of essays that cite medieval and early modern works and need to identify folios and columns, the abbreviations “fol.” and “col.” should be used and, for the sake of clarity, “p.” or “pp.” as well. Citation of a foliated volume, moreover, should recognize that the folio is a full leaf, having a front and a back, with the front or “recto” being the side on which the number appears, typically the side to the right of the gutter, and the back or “verso” being the side after the numbered side of the folio. Usual citation of folios will, therefore, read, for example, “fol. 32r” for the front or recto of folio 32, “fol. 32v” for the verso. There are also many early modern printings that either lack page or folio numbers entirely or lack them in the prefatory materials. In these cases, there will usually be printers’ marks at the bottom of the recto sides of at least the first several leaves in a quire of pages: e.g., “Ai” or “a*” followed by “Aii” or “a**” and so forth. In these cases, the printers’ marks are taken as foliation and the reference would read “fol. a* r” or “fol. a* v” — etc.

The availability of so much material on the Internet raises the issue of proper patterns of citation for works made available electronically either for viewing or for download. If the source to be cited is a PDF or GIF version of an extant hard-­copy, whether of a book or journal article, the simple solution is to offer a standard footnote to the author and title, and either place of publication, publisher, and date or to the journal location, without any indication that the work is a PDF or GIF download. For example, a citation such as “Thomas Halyburton, The Great Concern of Salvation, second edition (Glasgow: Robert Urie, 1751), p. 67,” would serve whether one was citing a copy from a rare book room, from Eighteenth-­Century Collections Online (ECCO), or from Google Books — inasmuch as a full bibliographical footnote would not normally indicate the library from which one had obtained a volume (citations of archival materials being the exception that proves the rule). Similarly, citations of a journal article from the original hard copy and from a database PDF version would be identical. This logic also applies to the case of on-­line journals that are not published in hard copy but follow the standard model of successive issues appearing annually or semi-­annually with a standardized page-­format. Thus, Anthony Dupont, “Continuity or Discontinuity in Augustine? Is there an ‘Early Augustine’ and What is his View of Grace,” in Ars Disputandi, 8 (2008), pp. 67-79, is a suitable citation, even though Ars Disputandi is a purely online journal in downloadable and printable PDF format. Note that in many libraries, the catalogue references to periodicals will include the URL as a life link in cases where the periodical is available online. In addition, many libraries have ceased to subscribe to the hard-­copy versions of journals that are available online in electronic form. In other words, written works that retain the standard identifiers including pagination, need not be cited differently when the mode of access varies.

This traditional bibliographical approach does not, of course, obtain for works available online as websites with unnumbered web pages, whether in the form of essays, encyclopedias, monographs, or digitalized text. Such works need to be cited as Internet sources, with the full “http://www . . .” URL, preferably giving the date of most recent editing or updating from the source on the web or, lacking that, the date of access, given the likelihood that the entry will change. Thus, e.g., Anonymous, “Philipp Melanchthon,” in Wikipedia, at https://en.wikipedia.org/wiki/Philip_Melanchthon, accessed 11/4/2015.

Above all, the form of initial and subsequent references should be standardized and maintained throughout an essay. Typically, a researcher will follow the patterns recommended by standard manuals of style such as The Chicago Manual of Style, or Kate Turabian, A Manual for Writers, or the SBL Handbook of Style. Various publishers supply their own manuals. All such works, however, provide only a beginning basis for footnote and bibliographical forms: researchers will find that not all of their referencing needs will be covered in any given manual. What is more, manuals — perhaps most notably the Chicago Manual of Style and its half-­sister, Turabian — are irritatingly inconsistent from edition to edition, making not only unnecessary but often unwise, barbaric, and arbitrary alterations of forms. By way of example: the time-­honored pattern for citation of multi-­volume works in a footnote is to place a comma after the title, note the number of volumes (“vols.”) and then to proceed with the publishing information within parentheses. One particular edition of the Chicago Manual altered this pattern without any justification, placing the volume referencing within the parentheses together with the publishing information — and in the very next edition reverted to the original pattern! The scholar must exercise judgment in such matters, strive for consistency, and simply ignore the current idiosyncrasies. A well-­constructed first reference footnote will read like a sentence, with the author as the subject, followed by the title as a predicate, and concluded by various specifications (volume number, journal issue, title and publishing information in the case of an essay included in a book, concluding with section reference or pagination). At least one of the authors of the present volume views the current practice of placing pagination after the title, followed by a period and a capitalized “In” to introduce further publishing information, as a significant loss of the grammar of citation and a descent into barbarism.

Bibliographies

An enormous bibliography at the end of a chapter that bears little relationship to the footnotes is unacceptable. On the other hand, a direct correlation between footnotes and the bibliography makes a separate bibliography for each chapter entirely gratuitous. The first footnote reference should be a full bibliographical reference; after that, the item will be referred to in a shortened form. As a rule, students should not add bibliographical references to the volumes that are cited in the footnotes (except in very rare circumstances).

The sole purpose of a bibliography in a dissertation is to provide readers with ready access to the sources used in the work without having to go through all the footnotes. Or it gives the reader the possibility of finding a full reference on the basis of a partial reference in a footnote, without going through the whole book to find the first reference. The bibliography provides a service primarily at that level. It is even justifiable to have a “Select Bibliography,” granting that some of the items cited are cited only once and may play only an ancillary role in the thesis; such works need not be included in the bibliography. Furthermore, it is a common practice today to provide a bibliography of manuscript sources alone, without a bibliography of secondary sources. If a full bibliography is required, it should be organized systematically under the two broad categories of primary and secondary sources. Under primary sources, one should list manuscripts and printed primary sources separately. Under secondary sources it is common to break a bibliography down further into categories of books, dissertations, and articles. Sometimes these three categories are run together alphabetically and sometimes they are not, but in either case, it is a good idea to work in this order and to think along these lines. Some dissertations may have bibliographical essays rather than bibliographies, but it is not common to utilize annotated bibliographies. Generally, the sort of annotation found in the latter should occur in the first paragraphs of each chapter, or in the introductory chapter. This type of analysis is part of the work of producing the monograph or dissertation and belongs in the text itself.

Research with Computers

Currently a wide range of computers and other forms of hardware with supporting software are utilized in graduate study, including a variety of word-­processing, text storage, text manipulation programs, optical character scanners, databases that assist in taking notes and in the formation of bibliographies, and statistical packages. This technology is ideal for research as well as for the actual production of a dissertation, article, or book. Each of these tools has its own technical vocabulary and some of them require highly specialized training and expertise. While in such short compass we will provide only the briefest of introductions to hardware options, leading technologies, and software packages, the references we include for further reading and exploration should be useful. Since all students will utilize a computer in the final production of their dissertation, we conclude with a separate, more detailed guide to word processing.

When one first sets out to purchase computer equipment, it is highly desirable to seek out the advice of a colleague or friend who has had some experience in the field. The technology in hardware is changing so rapidly that today’s advice may be utterly passé tomorrow. Nevertheless, a few general guidelines may be offered. In order to utilize the best available software to its fullest extent, students who purchase computers ought to think in terms of a medium-­ to high-­end PC or MAC with as much hard disk storage capacity as one can afford. The particular type of application, should, of course determine the exact configuration of hardware. A student working in the field of art history will require optimal capacity in storage, for example, whereas someone utilizing quantitative applications would naturally attend to the speed of the processor.

When recording research notes with a computer, the same general principles apply as in the practice of taking traditional, hard-­copy notes; a single unit of thought is separated as a distinct “file” from the preceding and the following materials. Units of data can be readily separated and identified and later organized by topic and subsection. Microsoft Office’s “One Note” and Corel’s digital notebook “Lightning” that is packaged with WordPerfect will admirably serve the task of note taking. In our experience, it has not been possible to avoid the use of hard-­copy notes altogether, but we concede that a new generation of students may be able to develop the skills that will enable them to carry out the entire process of research and writing without paper and pencil, especially in some areas of research, such as the editing of texts. Programs such as ProCite and EndNote (both by Thomson Reuters) offer the additional advantage of ordering bibliographic data into a database format that can then be rearranged and resorted in a great variety of ways. Citation, published by Oberon, facilitates both the organization of research notes and the online acquisition and formatting of bibliography. All three programs support the downloading of individual bibliographical references from national databases and the importing of these references directly into a bibliography that can then be automatically reformatted for different university style sheets, or transformed through the use of key-­words into footnote references that are readily inserted into word-­processed documents. The labor of retyping references for both footnotes and bibliographies is thereby greatly reduced, if not completely eliminated.

Hand-­held and full-­page scanners seem, on the surface, to offer significant promise for collecting data and taking notes, but they are fraught with no little danger for the beginning research student. Given that bibliographical references and vast quantities of original text, both primary and secondary, can be downloaded from library catalogs and other databases on the Internet and never pass through the keyboard of a computer, masses of reference data and quoted material can now be entered into students’ word processors without passing through their minds. The new technology thereby threatens to produce a fatal short circuit in the normal process of sifting and analysis, and because of this, it is a serious question whether or not the use of scanners and downloaded text will actually enhance a person’s work or save them time in the long run. Since the beginning research student is already tempted to overquote material, the use of a scanner and downloads may result in a deadly glut of unassimilated, and finally unusable information. The one obvious exception to this concern would be in the case of scanning an entire class of full documents, such as sermons, which would then be searched by the computer with a pre-­defined set of carefully formulated questions.

Most of the important databases that research students are likely to use, such as the Thesaurus Linguae Graecae or the English Short Title Catalog, have proprietary retrieval software that is written specifically for these databases. Some projects, however, may require the searching of texts and the use of statistical analyses in stand-­alone applications, and there are a number of excellent software programs that will greatly facilitate such projects. If a project entails the close analysis of words and phrases in several thousand pages of primary printed texts, then a text-­based management system will be required. Word processors, such as WordPerfect and Microsoft Word, have rudimentary forms of these programs, and words and phrases can be easily searched for and isolated. Students, however, should be aware that there are much faster and more efficient programs written specifically for this kind of task. The software known as “X1” by X1 Technologies finds and sorts text with much greater speed than standard word processors or the search engines embedded in operating systems (http://www.x1.com/). “Sonar Professional” is more sophisticated and supports virtually all forms of possible searches (Boolean, proximity, wildcard, synonym, and phonetic) and can search over 10,000 pages of text a second.5

For the organization and the manipulation of quantitative data, the PC or MAC version of the Statistical Package for the Social Sciences (SPSS) is an essential tool. This program will require some study to use effectively, but it does not necessitate the expertise of a statistician. If a student will master the pertinent sections of a basic text-­book, like Hubert M. Blalock, Jr.’s Social Statistics, they will soon be able to utilize the most important statistical procedures packaged with SPSS. Moreover, one of the greatest advantages of this highly sophisticated program is the assistance it offers in organizing and displaying quantitative data.6

Word Processing the Dissertation

It was not very long ago that the greatest agony suffered by doctoral students in moving toward the completion of the dissertation was the production of a fine, type-­written product. Either the student accepted the burden of typing the dissertation herself, learned the difficult technique of placing footnotes at the bottom of the page, and mastered all of the other niceties of formatting, and then re-­typing whole pages or even sub-­sections to accommodate corrections — or the student contracted with a typist for the work, and then still was responsible to see that the job was done properly. The personal computer has radically altered this situation and made possible the production of finely formatted and printed dissertations by students themselves. Indeed, the proper use of the word processor enables the integration of the work of research and writing with the production of the final printed form of the dissertation — and the adoption of a method of taking and collating notes on the computer such as that outlined above will greatly facilitate the transition between the early written draft of a dissertation and the final draft. Several issues need to be dealt with in this regard.

Students should acquaint themselves as quickly as possible with acceptable formats for footnotes, standard abbreviations, chapter headings and sub-­headings, tables of contents, and so forth. Much of the specific information needed on these matters is provided by such standard works as The Chicago Manual of Style and Turabian’s Manual for Writers of Term-­­Papers, Theses, and Dissertations, which is based on the Chicago Manual.

The best word processing software available today will provide students with such capabilities as (1) the creation of different fonts (italics, underline, bold, larger and smaller size type, superscripted footnote numbers, etc.), (2) automatic placement of footnotes at the bottom of a page, (3) automatic re-­numbering of footnotes, (4) generation of large (i.e., book or dissertation-­length), continuously paginated master documents from a series of shorter files, (5) the ability to code headings and subheadings for the purpose of generating a table of contents, (6) the ability to code headings, pages, and notes in order to create patterns of internal cross-­referencing automatically targeted to page numbers, (7) the ability to introduce running heads for chapters and to suppress the header on initial or title pages, (8) the ability to code text to avoid widowed and orphaned lines and to attach section headings to paragraphs to prevent separation at a page break, and so forth.

The selection of a particular word processing program for use in the writing of a dissertation should be made on the basis of a clear knowledge of the program’s capabilities. It is not our purpose or intention here to recommend any particular software package. We would strongly advise against any package, however, which is incapable of performing any one of the above noted functions; Microsoft Word, WordPerfect, and Nota Bene admirably satisfy all of these criteria. Microsoft Word has clearly become the dominant word processing program, but students and researchers need to be aware that other programs have features that are more directly adapted to the needs of researchers and writers of scholarly monographs. WordPerfect’s “reveal codes” function, manner of coding font changes, and highly adaptable approach to document styles give it a distinct advantage. Recent reviews give Microsoft Word, TextMaker, and CorelWrite equal usability ratings but place WordPerfect higher. There are also solidly rated free programs such as Apache OpenOffice. Nota Bene in particular is very efficient with editing short-­cuts, and the seamless integration between the text editor and the text database is probably superior to any other word processor (http://www.notabene.com/). Guides are typically included with these packages, but printed guides are also available and will offer an overview on such matters as sub-­programs, built-­in utilities, and macros that are not obvious to the uninitiated. In addition, students should be aware that major software programs are often available to students at considerable discounts.

Manuals and guides (whether in hard copy or online) will provide examples of how to code bibliographical references so that the text of the reference indents automatically below the first line of the reference — and does so in such a way that editing or changing the font size of the reference does not create any spacing or coding problems. If the first line of the reference is simply typed and then either indented or tabbed in for the second, the text will be unstable and any editing will destroy the neat formatting. These manuals also offer clear and useful advice for such WordPerfect features as “Macros,” which will be of great service in the use, for example, of short titles in footnotes.

The continuous use of a good software program like WordPerfect will become a self-­improving exercise, since the techniques that are required to improve the manuscript are readily mastered. The various niceties of WordPerfect or Microsoft Word can be learned easily and then used consistently without any great problem of forgetting the basic command-­structure. Students should keep pressing on the limits of their knowledge of techniques in all of the areas discussed in this chapter, since they are all so logical that learning is incremental and becomes almost second-­nature.

1. Many standard texts examine basic research skills and techniques. Two of the best are Norman F. Cantor and Richard I. Schneider, How to Study History (New York: Crowell, 1967); and especially, Jacques Barzun and Henry F. Graff, The Modern Researcher, 6th ed. (Belmont, CA: Thomson/Wadsworth, 2004). We will not dwell in this chapter on basic writing skills, but students are advised to read, and reread, William Strunk, Jr., and E. B. White, The Elements of Style, 4th ed. (New York: Longman, 2000); and William Zinsser, On Writing Well: An Informal Guide to Writing Nonfiction, 5th ed. (New York: HarperPerennial, 1994).

2. G. Kitson Clark, Guide for Research Students Working on Historical Subjects, 2nd ed. (Cambridge: Cambridge University Press, 1968), p. 19.

3. Barzun and Graff, Modern Researcher, p. 258.

4. Kitson Clark, Guide for Research Students, p. 34.

5. Sonar is produced by Virginia Systems Inc. and its main drawback is its cost. http://www.virginiasystems.com/.

6. See the guide in the appendix.