5 Making Source Notes

You should read with a pen in your hand and enter…short hints of what you feel…may be useful; for this be the best method of imprinting [them] in your memory. Benjamin Franklin (1706-1790)

To be able to play with ideas, we first have to liberate them from their original context by means of abstraction and respecification. Sönke Ahrens.

As I mentioned previously, sometimes people feel a sense of “writer’s block” at the beginning of a new project, but often that feeling comes from a misunderstanding. In his book How to Take Smart Notes, Sönke Ahrens criticizes writing teachers who encourage their students to “brainstorm” to come up with a topic for an essay or a research project. I think processes that include free association and a sense of informal openness to surprising combinations of ideas can be very useful, especially in group settings. However, I tend to agree with Ahrens that if I’ve already done the work of highlighting, making Source Notes, and then turning them into Point Notes, then coming up with a topic to write about ought to be the least of my concerns. The whole point of this note-making process is not only to provide ourselves with ideas we want to pursue, but to actually show us which ideas we are most interested in. We’ll see this process in action as we continue. Right now, let’s look a the first stage in the writing process: reviewing highlights and turning the best ideas we found into Source Notes.

Boiling important ideas we encounter down into a sentence or two is the key skill we learn in this process. Taking a complex narrative from literature, or a textbook chapter, or a primary source, or a lecture and being able to say, “This is what that was about” is a crucial step in the journey from hearing about the knowledge of others to creating our own.

Highlighting is something most of us have been doing for so long that it may seem obvious. But you may have noticed a difference between the way you may have been taught to interact with texts and the suggestions you have found here. This difference mostly revolves around a slight redefinition of our relationship with texts. Although some texts you may read may be pillars of their disciplines, or even of a wider community or culture, in the note-making we are practicing here, the point is not so much appreciation as information. We are not reading texts simply to experience their greatness, but to learn something. Often something quite specific, that answers a particular question we are pursuing. While it’s still possible to maintain an attitude of reverence for a particular text or for great writing in general, it’s sometimes helpful to tone down that emotion a bit when we approach texts to learn from them — especially if that feeling makes it more difficult to read with a critical eye.

Once we have read and highlighted, the next step is to go back and review the highlighted passages. Rather than just copying down a quotation of what the author said, we try to summarize it in our own words. This is the stage when I begin to paraphrase rather than quote, unless there’s a really compelling reason to quote. Paraphrasing is how I demonstrate (to myself, at this point) that I got the author’s point. It is also the beginning of my response: what I write down and how I describe it helps me nail down my interpretation of the author’s point. It’s the beginning of a conversation between me as the reader, the text’s author, and my ultimate reader.

Summarizing and Paraphrasing are very closely related in this process. Educational research has shown that when students restate something they have learned in their own words, the information is much likelier to move from their short-term to long-term memories. This is probably because the words or images they use to describe the new knowledge already have a home in their world-view. Connecting them to an existing idea in their understanding of the world seems to create places for the new ideas to occupy in the students’ memories.

For an academic perspective, consider Luhmann’s description (quoted in Ahrens’ book):

The best method is to take notes — not excerpts, but condensed, reformulated accounts of a text. Rewriting what was already written almost automatically trains one to shift the attention towards frames, categories, and patterns …conditions and assumptions.

Working in the style of Luhmann, we could use the note to comment on the structure of the author’s argument or her assumptions, as well as simply summarizing her findings.

Some people might be concerned that as we paraphrase what an author wrote, we run the risk of changing their original meaning. This is probably true, but it is rarely something we ought to worry about. Even quoting changes the original meaning of a passage, since it removes the words from their original context in the text. Quoting or paraphrasing passages selectively or out of context has certainly been used to subvert the meanings of texts throughout history. But, I would argue, rarely by accident.

Paraphrasing is no more suspect than quoting, as far as accurately reflecting a text goes. But the added value of paraphrasing is that we are frequently not pursuing the same question as the author whose writing we are making into notes. We may have found a piece of data or an argument that is relevant to our pursuit in a completely different way than it functioned in the text where we found it. Or, like Luhmann, we may be interested in deconstructing the text’s frames, categories, etc.

Another objection people have raised about paraphrasing is that it is a short step from plagiarism, or at least on a slippery slope on the way to it. This is another misunderstanding. We have already seen how nearly all the knowledge we have is inherited from others. We use countless pieces of what we call “general knowledge” every minute without citing the sources where we first encountered them (if we can even remember them!). But the type of knowledge we’re taking from sources and making our own in our notes is a bit more specific than this general knowledge. Its sources deserve to be cited, because they are often recent and identifiable. And they need to be cited, because the sources provide a context for our own ideas without which they would lose a lot of their meaning and value.

So it is a regular part of note-making that we have a bibliography attached to our system of notes, where we store a complete record of the sources we have engaged with. When we write a source note, we cite the source just as if we were placing a footnote in an article. Later, if we use the note in an article, the citation can become a footnote.

You’ll notice that in addition to the citation, this Source Note contains keywords. These are like tags in an online system; they describe the idea on the card, even if the keyword is not mentioned in the idea. Neither “students” nor “motivation” is mentioned on the card, but that is the context I chose for the idea, when I was thinking about where in my outline for this book it ought to go. As it turns out, I used this idea early in Chapter 1. I know this because of the card number on the top right, which corresponds to the last section of that chapter (the numerical range of the notes I used was 3.1.1.1/1 to 3.1.1.1/25).

This brings us to the often misunderstood and sometimes controversial issue of note numbering. The difference between many digital systems and paper is that when we use a physical note system we have to write an Index Note and keep track the address of this note somehow, so we can find it later. Each note is going to be filed in a drawer in a card catalog. But not randomly. It will be filed by its card number.

This is not necessarily the case in digital note-making systems. It is possible to enter a note in a word processing app or a notes app such as Obsidian without tagging it any way. Nearly all software today includes powerful search capabilities. For that matter, operating systems include tools such as Spotlight Search that will allow you to find files that include keywords you can remember. But what if the idea you want to associate a note with does not appear in the text of your note? You’ve already seen a couple of examples of my notes where this is the case.

In a digital system, keywords and tags are probably a minimal level of identification for your notes. Writing something down does you very little good, beyond helping to cement a thought in your long-term memory, if you can’t find the note when you need it. I have experienced the frustration that comes from knowing I have long ago written about a source or an idea that would be very useful to a problem I’m working on, but not being able to locate it. I have multi-terabyte external storage that I can query with nearly-instant full-text searches. There are still times when I can’t find what I’m looking for. Even my more recent note system, in a brand-new, purpose-built app, became a “black hole” of thousands of notes that I could not easily search and arrange. This was not the fault of the app. The problem was that I had not left myself a bread-crumb trail that would lead me back to the ideas I had entered, or show me how they were connected to other ideas I was already working with, in the system.

Luckily, this is not an impossible situation to rectify. I am gradually reviewing my big mountain of notes, chipping off individual rocks and pebbles; connecting some and throwing others away. It’s a time-consuming process and seems especially wasteful because I’m discarding notes I don’t think I need. I can’t help wondering how much farther ahead I might be if I had chosen the right notes and connected them as I made them, in real-time?

The error I made, that I want to warn you against, is sometimes called the “collector’s fallacy”. Having a well-stocked library is great. But it is not the same as making a system of notes that capture and elaborate your ideas. And the distance between a personal knowledge management app filled with clippings and highlights that I haven’t thought about is about as great, I’ve found, as if I had just highlighted a book and put it back on the shelf. It’s better than nothing. But not what I was hoping for. Not what the PKM gurus promised, and I imagine not what you are looking for if you have bothered to read this far.

There are two levels of identifying each and every thought you put into your system, that I urge you to use. The first is tagging each note with a keyword that identifies what the idea described in the note is about. The keywords should reflect topics that are of interest to you, rather than a predetermined list of categories like the Library of Congress catalog system. The note-cards that are contributing to this section of the chapter use the tag “Keywords”. This card, for example, describes the index that can show us where clusters of interest are concentrating. While it is valuable to have the keyword on the card as you’re shuffling through a small deck of them to arrange your thoughts, it is indispensable once you have more notes than you can spread across a big table and sort at once.

An Index is something you must physically create as you add cards in a physical note system. You create an alphabetized list, where you make a quick entry for each new note, under its keyword. This is where the card’s unique number shows its value. The Index card becomes a list of card numbers that can direct you to ideas throughout your system and show you connections between different trains of thought.

But why not just create a category called “Keywords” and file every card you write on that topic in that section? This might be all you need, if your notes are directed toward a small, immediate goal. If you are writing a research paper, for example, and you’re planning on throwing the notes away when you are done, then it may make sense to stick with just keywords. But what if the question or project you are working on is bigger? What if you are researching for a thesis, or studying for your profession, or pursuing a lifelong interest? Then you will probably find that ideas related to a particular “keyword” thought can come from different sources at different times. And, even more important, many ideas can be relevant to more than one train of thought. Often, the exciting, innovative ideas you’ll come up with will be interdisciplinary, or at least will involve combinations of ideas you encountered in different contexts.

This is not to say that you should put a half-dozen keywords on every card. Most will have one; a few will get a couple right away. Some will get another keyword later, when you have revisited the idea and considered it from a new angle. The unique card number is also incredibly valuable when you want to create a link between a new thought you are adding to the system and one particular other thought, rather than a category of thoughts. Another of the practices that is crucial to making useful, connected notes is identifying a specific note to file each new note “behind” in the numerical sequence. We’ll cover that in detail shortly; for now it’s enough to understand that each idea is a unique thought in the note system, and gets its own card, keyword, and number.

Numbering systems have also been a source of controversy in the PKM community. Some users of digital systems eschew numbers entirely, believing that the searchability of the system and the ability to create “Maps of Content” make numbering redundant and pointless. As someone who has seen my digital system become a mushy mess of hard-to-locate ideas, I disagree. Also, I have seen my ideas about how to organize my “Map” change over time; requiring time-consuming reorganization of my notes. Finally, although I had noticed clusters forming and had been able to see connections in “graph views” of my data, I was not satisfied with my ability to get information out of the digital system. It satisfied the “collector” in me, but not so much the writer.

The numbers themselves have also been a source of debate. Some digital users identify a new note chronologically. One I made right now, for example, might be numbered “202207201003”, which would be unique in my system, provided I don’t make another this minute. The advantage of this system is that I could keep track of when I had particular ideas, which might come in handy sometime in the future. The disadvantage is that the number doesn’t convey any additional information, and it doesn’t allow me to choose where to insert a new note “behind” the existing note it is most closely related to.

I’m going to briefly describe how I am currently numbering my notes. I do not claim that this is the only way that works, or that it’s the best system for you. I don’t even claim that it’s the system I’m going to stick with forever. But it works for me and you’re welcome to use it as a starting point, and then tweak it for your own needs as you become more familiar with your own unique process.

Developing a numbering scheme, I wanted to remain focused on getting to output, rather than trying to conform to any standards of correct taxonomy. For example, I don’t think it’s wrong just to begin numbering the first card “1” and the second one “2” and just carry on from there. The keywords and index are really the most important things here, to allow you to find and link your thoughts. Ultimately, the number of a card is not meaningful, except as a unique identity and as a way to locate it. So it would be perfectly okay to begin with “1” and “2”, and then when you have another thought that modifies “1”, put it behind that card and call it “1a”.

My own numbering system began as a reaction to a conversation with another analog note user who recommended using a top-level numbering scheme that classifies knowledge by disciplines. Another friend suggested that if we were going to begin creating a shared space where others could view our notes on the web and if everyone used a common system that had the same number for the same discipline, it might be helpful. I think both ideas have merit, but I decided not to base my private notes on standardized categories. I don’t want to create a bunch of empty “folders” I may never fill, and I prefer my “trees” of knowledge to grow organically and not be tied to a predetermined “trellis”.

So my numbering system has a couple of peculiarities. Firstly, it has two number series, separated by a divider. For example, 1.4.2.15/16a. The numbers on the left of the divider are nested categories, like folders. The number to the right belongs to the actual card that corresponds with a single idea. The idea may be a piece of evidence or an argument; that is, a Source Note or a Point Note. So the overall number tells me something about how the idea is related to other ideas (the nested “folders”) as well as where it fits in a conversation with other ideas (the sequence number and letter).

Rather than choosing a set of top-level domains that reflected disciplines or the Library of Congress or the Dewey Decimal system, I chose the three things I thought I would be making notes on, History, Science, and Knowledge Work. Later I added a fourth, Human Behavior. I numbered them 1 to 4. History is “1”. Within History, I made seven additional categories for the six continents and one for “World”. I started with Africa, because that’s where humans began. Within the fourth subdivision, which was North America, I used “1” for my Secular Radicals project and “2” for my Primary Source project. So now I have a Topic, “History of North American Secular Radicals” that has a number 1.4.1. Primary Sources are 1.4.2. In that topic I have fifteen “folders” for the fifteen weeks of my semester. In some of these topics I have dozens of files. None of this was pre-planned as I was setting up the system. I added the branches as I needed them.

So 1.4.2.15/16a means: 1: History, 4: North America, 2: Primary Sources, 15: Reconstruction, /16: Politics in the South, a: An additional thought I had about this source. The additional thought I had in the “a” note is the beginning of an argument I want to make about Southern politics during Reconstruction. This is a “Point Note”, where note 1.4.2.15/16 was more of a “Source Note” (although it had a bit of interpretation on it, too). This is an example of how I’m not particularly interested in what the note is called; what matters is how it contributes to building the output I’m working on (in this case a Primary Source reader for my US History course).

The two main ways I will be able to access the idea on this note are first, that it will be physically present in the card drawer, so when I go there looking for my primary sources about the Reconstruction period of American History, it will be right where I need it. The second way I could find it would be if I was looking specifically for notes on Southern Politics, I could find it in the Index by searching “Politics” or “American South”. On each of those Index cards it would be surrounded by other entries on those topics that might create meaningful new contexts for the ideas in the note.

A new note in an existing series is typically added at the end. But when we return to a particular thought in that series and decide we want to add a comment on that particular note, we can do that by “branching” with a letter. For example, if I returned to note 3.1.2.2/2 (the note on keywords in the photo above) in the future and had an additional thought I wanted to record to modify the idea, I could add a note 3.1.2.2/2a and insert it in front of 3.1.2.2/3. If I return again and want to modify 3.1.2.2/2a, I can alternate back to a number (3.1.2.2/2a1). If I have another new thought about 3.1.2.2/2, I can call that 3.1.2.2/2b. The point of this elaborate numbering is that I can have a conversation with myself regarding this train of thought, over time. Like a real conversation, each new statement doesn’t invalidate or erase the previous ones, but builds on them. Our ideas evolve over time, and even moreso when we are aware of and participate in that evolution.