Base image taken by the Automatic Professor Machine
Base image taken by the Automatic Professor Machine
We all have heard about the recent shrinking of attention spans1 The obvious consequence is that we can't engage with any kind of long-form content, especially in activities that involve learning. To ameliorate the situation, recently a new learning method appeared, which may make the best out of this situation: microlearning. It comes in the form of book summaries, apps that present you with the “fact/quote of the day”, Duolingo-style apps for philosphy, etc. But is microlearning a good solution?
In this article, I will first argue that the inability to process long-form content is, in fact, a problem. Then, I will argue that a specific subset of microlearning—book summaries, book excerpts, and quotes— is not only not a solution to the problem, but actually exacerbates the problem.
Microlearning is a recent term and there doesn't seem to be a single, clear-cut definition in the literature. The following definition, which I found in a recent review2, seems to be relatively accurate:
[...] generally speaking, microlearning may be described as “a form of e-learning delivered in small chunks, focused on delivering skill-based and just-in-time knowledge” (JIT) [...] JIT knowledge can be defined as “delivering the right knowledge at the right time”
But this definition is too broad, at least for our discussion. If I understand it correctly, a book summary falls under microlearning, but so does asking ChatGPT a very specific question like: “What year did Chopin die?”. Actually, even asking an online guitar coach (I mean a living person) a question seems to fall under it. This is the problem with such a broad definition: it includes too much for us to have a meaningful discussion. For example, I have a problem with the first form of knowledge delivery, but not with the last two.
So, let me try to constrain our discussion a bit. I will focus only on:
I will dub them reductive bites.3 Why I use this term will become apparent immediately. Also, I have a problem only when reductive bites are used as learning material. This will be assumed in the following discussion (e.g., think of an app that “teaches” you what Descartes said by presenting you the quote “Cogito, ergo sum”)
I argue that there is a characteristic that distinguishes reductive bites from e.g., asking ChatGPT a specific question. That characteristic is that reductive bites (attempt to) reduce an otherwise deep idea, concept, thesis, ideology, theory, opinion, argument, hypothesis, etc. into a concise, small, shallow bite. On the other hand, a specific question to ChatGPT receives a shallow answer (e.g., 1849) because the concept it is concerned with is shallow! There is no reduction going on because there is no need to reduce anything (although of course that doesn't mean the answer is correct).
The immediate question that pops up, then, is: what is the problem with such reductions? Nothing, unless such reductions are used for learning purposes! Then, they face the following three problems:
When the topic is deep, authors spend a significant amount of time to contextualize and scope their discussion, and considerable effort to clarify and explain their arguments. Their goal, of course, is their discussions to be taken as intended, in their full meaning. But, reductive bites systematically throw such work away. Take, for example, this excerpt from the book “Sex at the Margins”, by Laura María Agustín:
The journeys of women who work in the sex industry are treated as involuntary in a victimising discourse known as 'trafficking' [...] Migrants working in the informal sector are treated as passive subjects rather than as normal people looking for conventional opportunities, conditions and pleasures, who may prefer to sell sex to their other options.
Reading that, you may get a strong reaction and form a certain (possibly non-positive) opinion about the author/book. But, here's another excerpt from the book:
Sex at the Margins examines current ideas about this phenomenon of travel and work, demonstrating the discursive gaps and silences through which poorer and undocumented people slip, especially women who sell sex. Usually, these slippages are blamed on abstractions—society, the state— but this book argues that those declaring themselves to be helpers actively reproduce the marginalisation they condemn.
This now starts shedding light to the fact that this book is much deeper than it originally appeared, and that the original excerpt was probably misleading.
In summary, there is a reason authors spend time writing long paragraphs, sections, chapters, even series of books; it's because there is some depth they're dealing with. An excerpt can be useful as a reminder of the content, but only after one has actually read the original. In other words, it's bad learning material. We'll talk more about this issue when we talk about exformation.
If we get 10 people to read the same summary, they will probably all take away the same thing. Why? Because there's only one, shallow thing to take away. Even if the original content has depth, multiple facets, arguments and sub-arguments, digressions which may lead you to think of something interesting, etc., all those are reduced to a one-dimensional description.
Take, for example, the book “Civilization and its Discontents” by Sigmund Freud. Here's an excerpt from a summary of the book:
In general, the ego perceives itself as maintaining “sharp and clear lines of demarcation” with the outside world. This distinction between inside and outside is a crucial part of the process of psychological development, allowing the ego to recognize a “reality” separate from itself. After summarizing his previous research, Freud returns to the question of “oceanic” feeling, finding it unconvincing as an explanation of the source of religious sentiment in human beings. Instead, according to Freud, it is a longing for paternal protection in childhood that continues into adult life as a sustained “fear of the superior power of Fate.”
Most people would take away from this the same thing: “The book talks about the separation of the self from the outside world”.
But wait a minute! How does this demarcation happen? What is this “oceanic” feeling? Is there a difference between the physical and the abstract world for the ego?
It's only in the occasion that someone has actually read the book that they can offer a non-obvious insight like: “You know, Freud distinguishes between boundaries of the ego towards the inner and outer world. He mentions that it is known from past psycho-analytic research that when looking inwards towards one’s self, the limits of the ego are indistinct and blurry, but sharply and clearly outlined from the outer world”.
Aah! Now, that's interesting! But it's impossible to get such insights unless you read the book. We are heading towards more uniformity and less diversity. Who wants to hear for 100th time that Romeo and Juliet is about two families in animosity, or how great the balcony scene was? I prefer hearing about the nurse's husband's highly inappropriate jokes. But you probably won't find those in a summary.
Many times it appears that the microlearning culture is fueled by a dubious incentive for knowledge: people want to show off they know about something, rather than getting in touch with that work. As I wrote in a previous article, it's not about remembering, or knowing gists; it's about the unique impact a book leaves on you.
I think we may be losing touch of how moving and aesthetically pleasing one's unique writing style can be. It's another kind of uniformity; texts look more and more the same. That is probably partly due to everyone copying from ChatGPT and partly due to the unavoidable language leveling of political correctness. But also, I argue, it is partly because no one experiences the unique styles that do exist, so that they can get inspired from them!
I'll be a bit absolute here, but frankly there is just no way one can get a feel of Nietzsche's style by reading a summary. Even translations can barely do justice, let alone 1-page highlight reels! Same for Karl Marx, the Bible, Orwell, Gandhi, Rawls, Alan Turing, Malcolm X, Kundera, and Thomas Kuhn.
A writing style is content! If we don't read the unique style of the originals, we don't value that content; we don't develop through inspirations, and we don't incentivize the authors to express their unique styles.
In the next 2 sections, we will consider some insights from the book The User Illusion. One such insight is the bandwidth of consciousness. Although the senses provide the brain with about 11 million bits of information per second, consciousness processes only about 16-20 bits/s (with an upper bound of about 40-50). We will leave aside how staggering this difference is to make this fact relevant to our discussion: consciousness processes reductive bites! Every second, 11 million bits are reduced to 16-20-bit bites.
So, one may ask, if consciousness is having such a blast with just 20 bits/s, how can summaries not be enough for us? I will raise two points. First, in the case of consciousness, we don't have the option to process more per second. But when it comes to books, we do have the option to process more; so why not take it?
The second point is more subtle. Who creates the reductive bites for consciousness to consume? Clearly, it's not consciousness; you are not conscious that there's a reduction going on every second. It turns out that it's the subconscious that does the job. I know, shocking. But, there's something more subtle: your subconscious is different than mine. This means that if we're both looking at the same situation, we'll perceive it differently. This attacks at least two of the problems mentioned above. For one, we avoid uniformity because what I process is different than what you process. We can also think about the writing style. There are two perspectives here; one is to think about the writing style of... reality. Well, we do perceive this original writing style and not someone's summary. The other perspective is of course that we do perceive someone else's writing style; that of our subconscious. But, in this case, we can't do otherwise anyway, and the writing style of my subconscious is different than yours.
Another unique contribution of The User Illusion is the concept of exformation. When we communicate, we exchange information (e.g., the words we speak). But, in doing so, we explicitly discard information that is in our heads. Tor Nørretranders, the author of The User Illusion, calls this “exformation”. In general, the process goes as follows: we have a huge structure of information in our heads. Before we speak, we condense certain paths of this structure to a couple of words which we communicate. These words enter the head of the other person and the excite certain paths in the structure of information in their heads.
Good communication happens when the exformation in my head excites similar exformation in your head. The more “in tune” we are, the more in sync our exformation is. Also, the more “in tune” we are, the less information we need to communicate. This insight helps us see through a different lens certain communication instances. For example, in The User Illusion the author mentions “the shortest correspondence ever recorded”; that between Victor Hugo and his publisher. After publishing Les Misérables, Hugo wanted to know if the sales are going well. So, he sent to his publisher a letter that only included “?”. His publisher replied “!”, which meant “things are going great”. Obviously, the information exchanged was incredibly impoverished; almost nothing if we express it in bits. But, because the two men were so much in tune, not only did the correspondence make perfect sense to both, but also it created copious amounts of synced exformation in the men's minds.
The same basis can help us understand why the same reductive bite is beneficial in some situations but negative in others; it depends on the exformation it excites. Usually, a reductive bite excites unique exformation when we have already consumed the original content. That's a why a summary can be useful in reminding you certain insights you got from the book. And that's why your tailored-to-your-reading-experience summary will probably be more useful (to you) than a generic summary. This is also why context is key! Books usually spend considerable effort to set the stage and bring in your mind the context that is in the author's mind. This, in turn, helps to later excite more and more unique exformation in later parts. A reductive bite simply doesn't have the capacity to do that.
I'll finish this article with a couple of categories which may seem to be reductive bites, but I will argue this is not the case. My main counterpoint will be that for something to be a reductive bite, it needs to be learning material, and none of these are.
Most papers and academic theses have an abstract which summarizes the work. This is not microlearning simply because no serious researcher considers the abstract learning material. People usually read abstracts to see if the work is interesting or relevant. If they want to learn, they read the work.
Book reviews are not learning resources either. Again, they just try to entice the reader to read the book.
Here's a good way to think about how criminally misleading things like book summaries and excerpts are. Read (a part of) any constitution, and then tell a lawyer that you know anything of real value about that constitution. You will probably be met with a seriously questioning look.
I'm not a lawyer, but here's my 2 cents from my humbling experience. Last year, I developed an interest for constitutional law. After reading long parts of the Greek constitution, and reading textbooks on constitutional law, I learned one thing: I pretty much don't know anything about constitutional law, and I won't get any valuable knowledge about it any time soon. A constitution is basically useful only to people who have spent years analyzing it, studying law and precedent cases, understanding subtle terms and complex footnote reference graphs, etc. It is not of any value, at least on its own, to me.
A similar story is true for definitions of all sorts. Here's, for example, a definition of populism from Cas Mudde's book “Populism, A Very Short Introduction”:
[Populism is] a thin-centered ideology that considers society to be ultimately separated into two homogeneous and antagonistic camps, “the pure people” versus “the corrupt elite”, and which argues that politics should be an expression of the volonté générale (general will) of the people.
After reading it, I was like “I think I get it”. And no, I didn't. Cas Mudde spends quite a bit of space to explain what he means by “thin-centered”, “ideology”, “the people”, “the elite”, “the general will”. This definition is actually pretty useful once you understand these concepts, because it defines populism in a concise manner. But it is at best misleading if we consider a learning resource on its own. To give you a taste of what I mean, did you notice that the definition says “volonté générale”? Why didn't it just say “general will”? It anyway seems that “volonté générale” and “general will” are the same thing, just that one is a translation of the other. Well, no. “Volonté générale” has a long history and a hefty weight, that dates back to Rousseau and the Declaration of Human Rights.
This is basically how summaries are, as I said above in the exformation section. Summaries of papers or books I've already spent time on are very useful to me because they remind of what this book was about, which then reminds me of my insights. But, in and of itself, it's more or less useless, because I don't know the context and I won't develop any unique insights from it.
Shorter versions of texts usually appear in academic works, where we have the main the submitted and published paper, and an extended version.4 Another possible format is that used by Ray Dalio in his book “Principles for Dealing with the Changing World Order: Why Nations Succeed or Fail”. It is written such that it combines a short version, which is bolded, and a long version, which is the full book.
Short versions of works are still not reductive bites because, while they are learning material, they are usually not that short to oversimplify. For example, one can get a comprehensive and accurate idea of a paper,5 with enough background and references to prior work, without reading the extended version. Similarly, I can tell from experience that you can definitely get a deep appreciate of Ray Dalio's arguments without reading the full book.
Neil Postman wrote the incredible Amusing Ourselves to Death: Public Discourse in the Age of Show Business; it's short, to the point, and thought-provoking. In that seminal book, Neil Postman claims that: (a) a previously print-culture has now turned to a TV-culture, (b) that as a result society has come to value entertainment much more than exposition, and (c) that this is a clear regression. Postman focuses on the medium—text vs TV—similar to McLuhan's aphorism “the medium is the message”. Pretty much everything in the book is true, or truer, in the Internet age of today. A big reason microlearning appeared is that the media :Internet, computers, and smartphones promote this kind of learning (which makes Chapter 10 of the book especially relevant).