My grandfather's research technique - Index, Proto-Zettelkasten, Paper

My grandfather's way of doing history research before the computer age. Using paper and a meticulous process, he built something close to a Zettelkasten.

My grandfather's research technique - Index, Proto-Zettelkasten, Paper
Photo illustration by Cortex Futura. Original via (lorem ipsum)

My father recently sent me a study/research-guide that my grandfather had printed in 1984 for his students called "Introduction to the Study of Eastern Church History". He was a history professor researching Eastern Christianity and wrote his dissertation about "The Jacobite Church in the Age of the Syrian Renaissance".
I never met him, but judging from stories and evidenced by this text he was a meticulous researcher and took his work very seriously.

Flipping through this short guide, one chapter in particular caught my attention: "Arbeitstechnik" [Work technique].
Reading through it, I found it remarkable how much of his technique was similar (although not identical) to Niklas Luhmann's concept of a Zettelkasten - in fact, my grandfather uses the word "verzetteln" to describe his technique in one place. I don't know that my grandfather had contact with or knew of Luhmann - Luhmann's seminal text "Communicating with Slip Boxes" was published in 1992, and my grandfather apparently had used this system in the 50s already. His system is also not exactly a Zettelkasten: excerpts aren't linked with a continuous numbering scheme for one, and also aren't explicitly connected atomically through cross-references. That being said, he does employ something that reads as something close to what block-references are in Roam Research.

I also found it striking just how analog his system was. He wrote his excerpts by hand, and manually linked them through a paper index, which he then used to compile chapters on entries in the index. We live in a much easier world now, don't we?

On Twitter I asked whether anyone would be interested in reading a translation of this chapter, and interest seemed substantial enough to actually do the work to copy and translate it. The bulk of the translation was done by DeepL, which I found incredibly good -- all I had to do in most places was correct one or two words to clarify meaning. Getting the text off the page into a digital format was actually way more work than the translation itself.

Most likely I'll also make the original text and the translation available as a page in a public Roam database. But since I'm not sure yet how to best do this, that'll take a while longer. I'll add a link here once that is done.

For now, here's the translation into English:

Peter Kawerau - Introduction to the Study of Eastern Church History

6. Work technique

The technique I recommend here, which I've often tested and used for all my works, which grants a high measure of certainty, speed and delivery on schedule in carrying out a scientific work, makes in the beginning for pretty tedious note taking, but proves to be fast and reliable if you do it diligently.

For that matter the sentence holds: he who wants to write something has to write, write a lot. If you can write shorthand which you are certain you can still read after a long time, you may use it. I've written everything in longhand myself. This technique also has the advantage that you can interrupt it at any time, be it due to an illness, to take a vacation or what ever other reasons necessitate an interruption. You close the lid, carefully stow the work away, and continue without any difficulties at the same point at which you left off.

First a word about the working material: You need concept paper (80g satin), which can be written on both sides with ink, preferably in the size DIN A4 (210 x 297mm), plus typewriter paper for the later fair copy (machine type) in the same size. This includes an envelope (Juris folder) for storing the collection of material, later another one for storing the typed work. Since a scientific work can take years to complete, it is important to make sure that storage is dust-proof from the outset. A sturdy card index box (preferably made of wood) is needed for the literature collection, along with a suitable index for alphabetical collection and classification of literature. Which DIN format to choose depends on the size of your own handwriting: if you have a small, delicate handwriting, you can get by with a smaller format; if you write with a broomstick, you need a larger box and larger index. This is of course also a question of money. But it's better to buy the right, expensive material at the beginning than to spend years making everything unnecessarily difficult with unsuitable materials. For the index cards you also choose 80g satin paper, no cardboard, because the thicker the index card is, the more space it needs in the index box - so no cards as you are used to from the catalogues of the libraries, but good writing paper in DIN format of the index box. Whether someone writes with ink, ballpoint pen or pencil is up to him, ink is most durable, pencil blurs easily over time and often cannot be read accurately. A green and a red ballpoint pen are needed for the later steps. In addition, a bunch of blue or red rulers, about 15cm long, which stand out from the white book and writing paper, as you can get them in any department store. They are an indispensable aid for the eye when copying source texts.

Work equipment and tools

Before we begin our work, i.e. the study of sources and literature and their excerpts for the purpose of collecting material, we will acquire a few simple but indispensable tools:

  1. The Index. It is marked at the top with the title of the work and then with the alphabet written consecutively: Here we will later enter the keywords, in alphabetical order, under which we will write the excerpted texts. Therefore, the letters at the very left margin are written in CAPITAL letters and next to them there is plenty of space to write the excerpted keywords. The index of keywords may only be written on the front side. During the excerpting process, the index must be kept next to the excerpt in order to quickly find already existing keywords, under which a new excerpt belongs, and to avoid double keywords that are factually close to each other. An example of such a keyword index is attached to this introduction and its function will be explained in detail.

  2. The list of abbreviations. This is where we enter the abbreviations we use for the sources and literature we use; this saves us having to give the title of the work in every excerpt, but only the abbreviation. The abbreviations will be listed later in alphabetical order at the end of the work and will be accompanied by the corresponding number of the bibliography. This means that the title of the work - now, however, quite exact and complete - needs to be written out only once, namely in the bibliography at the end of the work.

  3. The bibliography. We title the sheet with "bibliography" and number the sheets consecutively with 1, 2,3 etc., depending on the length of the bibliography, in the upper right-hand corner. In this bibliography, the books we actually use are entered with consecutive numbers, and the number is written after the abbreviation we give the book for our work - usually the name of the author or the first letters of the source - followed by the abbreviated title with author, and finally, on the far right, the signature of the library from which we borrowed the book and used it. This is not only an additional check, but also necessary in case we need the book again later, when we have already returned it. Then, when ordering an inter-library loan, the borrowing library and the signature of that library must be noted immediately on the order form, which speeds up the procurement process considerably. If the book is from our own library, we note "own". The bibliography must be strictly distinguished from the alphabetical index in our card index box. In this box we also put notes with titles that we have found somewhere in the meantime as necessary for our work, so the card index in the box is at the same time a literature collection, while the bibliography contains only the literature we actually use, plus of course the exact titles of the literature we have used in the meantime.

Before we now get down to work, let us familiarize ourselves well with a quotation from Kant's Critique of Pure Reason, Introduction to the 2nd Book of Transcendental Analysis (ed. Kehrbach pp. 139-140) Kant speaks here of that which is the unconditional prerequisite for every scientific work, JUDGEMENT:

"If the understanding in general is explained as the faculty of rules, then the power of judgment is the faculty of subsuming under rules, i.e., of determining whether something stands under a given rule (casus datae legis) or not. ... and so it becomes clear that although the understanding is certainly capable of being instructed and equipped through rules, the power of judgment is a special talent that cannot be taught but only practiced. ... The lack of the power of judgment is that which is properly called stupidity, and such a failing is not to be helped."

Thus, if one does not possess the power of judgement, that is, the ability to subsume under rules, the procedure described in the following cannot be used.

The index is a collection of vocabulary which is not only taken from the text and arranged alphabetically, but the first, still rough application of the power of judgement to our sources. The rule under which the keywords are to be subsumed is the subject of our work, which we have written above it as a headline in the first working step. With each sentence of the sources, which we have to read like a hunting dog, we ask ourselves: Does the text belong under our topic "The Jacobite Church in the Age of the Syrian Renaissance" or not? If so, then we will think of a noun that expresses as a keyword the essential content of the text. We enter this keyword in its alphabetical place in the index, take the first sheet of concept paper and copy the sentence. Before that we fold the sheet in the middle and write the source text always on the right half. The left half remains free for notes, literature references, factual information or even your own flashes of inspiration. Do not save paper under any circumstances!

At the top left we write the first letter of the keyword to make it easier to keep the accumulating pages in order, at the top right we write continuously on the front the numbers 1, 2, 3, etc. of the leaves of the keyword, which we want to call a on the front and b on the back; it is enough to remember this as a silent agreement.

Often our excerpt will contain more than the extracted keyword says. This does no harm; such things will happen all the time. What goes beyond the keyword will be transferred to other keywords in a second work process until everything that sources and literature provide is exactly under the keyword to which it belongs. It may happen that the same excerpt is completely or partially under three, four or more keywords. This is normal and must be so.

Once we have completely evaluated the sources and literature, we put our collection of concept sheets in alphabetical order by keywords and start on sheet A1 ... with the second work step, until we have arrived on sheet Z x ... .

We read through the excerpts again like a hunting dog and consider whether the keyword above, which is now the rule, should be subsumed under that, or whether the excerpt belongs completely or - which will usually be the case - partially under another keyword. We skim the index of keywords next to us, whether a suitable keyword already exists or whether a new keyword must be ejected. Depending on this, the part of the excerpt that is not correct is transferred to the new note.

To keep the material in your hand and not lose control over it, we take the red ballpoint pen, enclose the part of the excerpt that is to be transferred to another keyword in round brackets or, if these are already taken, in square brackets or angle brackets, repeat these on the left margin and write the keyword with the page number and a or b (front or back). Then we take the new keyword, write the old keyword with page number + a or b. in red (to avoid confusion with text and quotation) and then follow the text of the excerpt including the source document with blue ink.

In this way we work through keyword by keyword, constantly testing our power of judgement to see if we have judged correctly and attentively. We do this up to the last page Zb; then the second work step is finished.

Before we discuss the third and most difficult one, i.e. the preparation of the manuscript, let us take an example from my work "The Jacobite Church in the Age of the Syrian Renaissance. Idea and reality. With 3 illustrations and 1 map. Second, supplemented edition. Berlin: Akadmie-Verlag 1960 (Berliner Byzantinistische Arbeiten 3.) show how this is done practically.

As a small help, which will not always and not in all points be applicable, first a few aspects, which you will find useful from time to time, if you want to get "from text to dissertation":

  1. exegesis of the text Textkritik. - Übersetzung bzw. Heranziehen aller vorhandenen Übersetzungen. - Gliederung und Zusammenhang des Textes. - Erklärung einzelner Wörter und Begriffe. - Grammatikalische Erklärungen. - Individueller SPrachgebrauch des Autors. - Kirchengeschichtliche Vergleiche. - Literarkritische, formgeschichtliche oder theologische Untersuchungen.

  2. rules for the structure of a scientific paper These rules will be discussed below in a special section on the structure of a paper.

The text I am discussing for simplicity in the Latin version is in the 2nd part of the Chronicon Ecclesiasticum of the Maphrian Bar Hebrew (1226-1286) page 432; we are thus in the middle of the Mongolian period. Part 1 of the Chronicon Ecclesiasticum deals with the patriarchs of the Western Syrian Jacobite (monophytic) Church, whose nominal seat was Antioachia on the Orontes in Syria. Part 2 deals with the series of the Maphrians, what the heads of the Western Syrian Monophysites, who originally lived outside the Imperium Romanum, were called, which are given in Latin as Primates Orientis, as the title on page 433-434, top, shows.

We begin page 432 in the upper left-hand corner at number 62: Post Ignatium qui et Saliba, Gregorius seu Abulpharagius Bar-Aaron. What does this mean for our work and, first of all, for our index? Already conspicuous is "qui et Saliba", who is also (called) Saliba. Why this double name? Answer: Primates or Maphrians also took an official name when they took office. So we note these words in the index under A Amtsname [O Office name] 1 a. Was it always like this? Was this Maphrian Ignatius perhaps an exception? We can decide that only after we have completely worked through the Chronicon Ecclesiasticum. As abbreviation I have chosen in my dissertation in the list of abbreviations KG (=Church History) with reference to Lit. verz. No. (22); on page 128 the reader finds under No. 22 the complete title of the Chronicon Ecclesiasticum under its author Bar Hebrew. This is the only place where it is necessary to give the title with the utmost accuracy and completeness.

The indicated successor of Ignatius is our Bar Hebrew (Bar-Aaron), who had the official name Gregorius. Useful is the indication of the year 1264 A.D., in which Bar Hebrew took his office as Maphrian, which we find in numerous directories. In the margin we note: "List of Patriarchs and Maphrians with reign and civil name!"

Next: "Seu Abulpharagius Bar-Aaron". Striking and worth noting is the arabicising, islamic name form Abulfragius (Abu l-Farag). We note on a new sheet under the keyword "Islamic name forms" this passage, which shows us that Christian clergymen could also bear names after Arabic-Islamic forms; Saliba with Ignatius was a genuine Christian name in Syrian. Bar-Aaron, a Syrian-Hebrew name form, could indicate that the father of Bar Hebrew was still a Jew and that only Bar Hebrew converted to Christianity. We note this fact on a new piece of paper under the keyword "Judaism".

Then we read on: "Eodem tempore quo obiit Ignatius maphrianus (+1258 A.D.), qui et Saliba, perturbatae sunt regiones, et vastatum fuit Baghdadum, ac post eam civitatem pariter vastatum Alepum cum universa Syria et Mesopotamia;" concerns the Mongol Wars, 1248 A. D. Hügälü, the founder of the Mongol empire of the Ilkhane in Persia, the former capital of the Islamic caliphal empire of the Abbasids, conquered Baghdad. Aleppo and Syria and Mesopotamia were also devastated by the Mongols. It is striking that the Mongols are not mentioned with a single syllable, while the devastations mentioned afterwards are attributed to the Arabs by name. This results in a clear pro-Mongolian tendency of Bar Hebrew, which we note under "Tendency of Bar Hebrew" or "Mongol-friendliness of Christians" or a similar keyword. For what were the reasons for this? This will become apparent in the course of the work.

In a good history atlas for the Middle Ages, for example the Great Historical World Atlas. , published by Bayerischer Schulbuchverlag 2nd part: Middle Ages. Munich 1970 we inform ourselves about the situation of Baghdad, Ale 0, Syria and Mesopotamia. So while the Mongols are not mentioned with a word as the originators of these devastations and the murder of the last Abbasid caliph (not mentioned here), Bar Hebrew continues: "post haed autem Arabes Assyriae et Ninives in Christianos illarum regionum invaserunt eosque comminuerunt;" The Arabs, that is, the Muslims, organized a persecution of Christians as revenge. We note it down on a new keyword "Persecutions of Christians", Bar Hebrew has no timidity towards the Muslims to call a spade a spade. The Mongols, in order to avenge or protect the Christians, strike back: "denique paulo post supervenere Tatari (=Mongols) qui Arabes occiderunt, atque ita communis clades contigit omnibus, mansitque ecclesia profligata in Oriente profligata vidua sex annorum spatio;" in the general confusion the church remained a widow, a new head could not be elected, so a vacancy of the Maphrianate occurred from 1258 to 1264: six years, as Bar Hebrew states exactly. We note down a keyword "vacancy" and we will immediately see that this was not the only vacancy. Question: Where did the Patriarch and the Maphrian reside? This will be noted in the margin and clarified in the course of the work. But this may be enough as a sample of how to "paper-slip" a source text.

Our keyword index will therefore look like this until now:

Figure 7: The keyword index

If you have several slips of paper for a book, they are always inscribed on the front page and numbered with a consecutive number in the upper left corner. If you have actually held the book in your hand, so that the title is based on your own title recording, you make a red angle at the top right. You work through the secondary literature in the same way as the sources and will very soon realize what it is worth and whether you should study it more closely or take a cursory note of it.

Since you can only read and work through one book at a time, interruptions do not disturb the progress of your work. You can see from the bibliography on which you last worked, and if you put a note in this book with the date and page where and when you stopped working on it, you can continue in a few minutes after returning exactly where you left off a long or short time ago. In general, regular dating in the text, in the slips of paper or on the cover is highly recommended. Beginning of work, interruptions (with indication of the reason), beginning of the preparation of the manuscript should be noted in any case. Apart from the moral stimulus that this gives, they give you a sense of how long such work takes. Also on the cards of the card index, you note when you ordered a book, when you received it, when you handed it in. This can be valuable for a number of reasons.

Related Topics

Join the Cortex Futura Newsletter

Subscribe below to receive free weekly emails with my best new content, or follow me on Twitter or YouTube.

Join and receive my best ideas on academic workflows, learning, and Roam Research