Vulnerability and lacking literacy in (academic) online communication

Many people who interact on so-called social media express confusion about the behaviour of their interlocutors. Besides the on-going friendly chatter, there is a number of “polarizing” issues in the context of which a single “like” or a couple of lines might spiral out of control. While there are certain names – like Rowling or Kimmel – whose mere mention might be taken as tantamount to “taking a side” and hence be causing outrage, there is a vast number of smaller occurences or patterns that might be less charged but equally perplexing. I’m thinking of unexpected outbursts or insults or even more low-key issues such as inappropriate forms of address. What’s going wrong in such situations? While there are many helpful explanations of specific forms of communicative failures, I’m still thinking there might be a fairly general problem related to the way we engage as readers (and writers) in such situations. As I see it, we commonly underestimate the differences between spoken and written language. Let me explain.

Online communication consists neither solely in spoken nor solely in written language. Rather, it yields a hybrid sometimes called fingered speech. This hybrid is at once attention-grabbing, like most spoken language directed at a definite interlocutor or audience, and randomly targeting whoever cares to pay attention, like much written language (messages to specific people aside). In a spoken exhange, the speaker commonly starts the conversation by making the topic relevant to the hearer. (Deidre Wilson and Dan Sperber have worked this out in detail.) I might, for instance, tell you that it’s raining. But I wouldn’t do it just so. Normally, I would do that only if I knew that you’re intending to go out or that it’s relevant to you in some other way. Not so in written texts. A treatise or a news report on a subject matter is of course targeting an interested audience, but normally it isn’t targeting anyone in particular. By contrast, much online communication has traits of both: the written form in an open forum comes with the authority of relative permanence, visibility, references, perhaps comments, and targets no one in particular; yet, many features of online posts, like the (seeming) spontaneiety, speed, informal tone, specific references etc. make it seem much more context-dependent and even personal. Seemingly, there is “someone like you”, i.e. someone scrolling on their device and adding to “the conversation”.

For those (like me) socialized with written and spoken language as distinctive forms, this creates a paradox: If you’re scrolling, the items you see are signalling relevance (as in spoken conversation) but they are usually not directed at you personally (as in many kinds of written text or adverts). Being socialized with news outlets and advertising, this is familiar to most of us, but online communication adds an extra layer because many social media posts have properties that are typical of highly context-dependent and personal communication. This paradox runs deeper than might meet the eye. On the one hand, it seems to be the context-dependence and personal features that draw us in. On the other hand, the written online text comes with special affordances such as likes or favs and comments that signal status, the relation to certain in-groups or out-groups (Alex Davies calls this bystander information.) and hence might be welcoming or alienating, independently of what is said in a particular post. This paradox also has a political dimension: On the one hand, the illusion of fairly personal communication seems to create a safe space in which you can safely lurk or reach out. On the other hand, this “safe space” allows us to be monitored by stakeholders in often undesirable ways. Despite these facts, the interactions take place in the (seeming) privacy of our own homes or phones, where communication seems targeted, intentional and personal.

If this is correct, much of our confusion resides in the vastly different expectations that we associate with written versus spoken language. Depending on the further development of the related technologies, many of the confusions in online communication might pass sooner or later, but for the time being, the hybrid of fingered speech confronts at least some of us with a lacking literacy. Approching such speech with the sensibilities of diligent academic readers, we will forever bemoan the inconsistency and impertinence of our interlocutors who fail to live up to the pertinent expectations. Approaching such speech with the in-group sensibilities of attentive listeners, we will forever bemoan the stiffness and ignorance (not to speak of moral failures) of out-group posters. Online communication is not one or the other and hence not either written or spoken language. Accordingly, we will be vulnerable to incoherent expectations as long as we keep mistaking the kind of language we’re confronted with for something else.

Doing philosophy …

How to start working systematically. From exploring to writing

Today was the final session of my three-day course on medieval philosophy. As so often when I’m about to send my students home after difficult work on texts, I worry that they end up being mainly confused and at a loss when it comes to writing their papers. So today my doubts got the better of me and I talked them through a brief sheet I had prepared on the fly. The main idea was to hand out questions or tasks that they can use to work through the material in order to get from a mainly explorative phase to actually start a systematic writing process. After talking them through the sheet, I asked them to come up with a response to each item and sketch a paper or thesis based on it. After 20 minutes preparation time, we discussed three sets of replies. The point was not to give exhaustive responses, but to mark starting points. My hope was that the process could be facilitated by indicating (in brackets) what they actually had to do. The results left me stunned. Below, I’ll just leave you with the notes on the sheet.

Getting started

Throughout the course, I asked students to come up with what I call structured questions targeting a difficult paragraph or phrase in the text. So the starting point for the exercise is to look for a difficult passage and write down a structured question about it. Going from there, students had to pick a passage and work on the following items.

Preliminary questions (mostly settled during the course)

What type of text is it?
What genre is it?
What about the text’s transmission / kind of edition / translation?
What kind of speech act is it? Who is the author? Who is the intended / targeted audience?

Operationalizing the question about the text:

Terminology: Identify problematic words or phrases. Try to clarify them. Are they technical terms or everyday expressions? (Here, you should do searches in encyclopedias and such like.)
Content: What would be the case if the claim were true? What would be the case if the claim were not true? If the opposite were true? (Here, you should mainly think through the claims and counterfactuals. Use your imagination.)
Context: What points of contact are there to other texts? Both terminological or content-related. (Browse through neighbouring passages, other texts, histories.)
Methodological aspects: Ask yourself honestly: What assumptions am I making? Which of my assumptions facilitate understanding? Which of my assumptions make the content of the text appear difficult, absurd, or false? – Which of my assumptions are directly supported by the text? – Which are not? – Qualify the way in which these assumptions are supported / not supported by the text. (Here, you try to evaluate your responses to the former items, i.e. the relation between your assumptions and the material, and figure out what kind of methodological demands they make on your arguments.)

Questions for your planned work:

Which of these points are addressed in secondary literature? (Checking this helps you figuring out for which items you’ll have to argue yourself. Most issues can be settled by referencing literature. Try to have your work focusing mainly on one of the above items.)
Which points are not sufficiently covered? (research gap)
In which respects would you like to diverge from or expand upon the literature? (your own contribution)

Misguided expectations in reading: texts as arguments

Imagine a tulip! Now let me tell you that every tulip is a flower. – What is the difference between these two previous sentences? Among the many differences, the first invites you, diligent reader, to do something, whereas the second does not. Arguably, the first sentence is somehow incomplete without your cooperation. You have to do some imagining and it is not set in stone that what you do is entirely predictable to me, the author of that imperative clause. For one thing, I can’t anticipate the colour of your imagined tulip. Although I’m almost certain that you imagined a coloured tulip, not a colourless one. In what follows, I’d like to suggest that certain (philosophical) texts demand our cooperation. But they do this not just in the sense that our reading might be less boring if we actively think along. Rather, our thoughts and experiences are part of the argument or project. I think this suggestion matters greatly in that this cooperative feature of texts is often ignored, especially in early modern texts. Hence, the way early modern arguments work is often misunderstood.

As some you know, I’ve just co-organised a workshop on reading as a social practice.* While my head is brimming with thoughts of my interlocutors that I have yet to work through, I’ll try to wind down by following up on an issue that had occupied my mind since I began pondering on my Socializing Minds and was brought to light again especially when listening to a brilliant talk by Dana Jalobeanu on “Interactive Reading of Francis Bacon’s Novum Organum” (you can find the abstract here). According to her, Bacon’s

“Novum Organum was actively read in a distinctive, collaborative manner that requires careful reconstruction. Focusing on the early Royal Society, I show that some of the virtuosi pooled their intellectual resources to decipher and interpret Bacon’s text. Their reading practices were not solitary acts of comprehension, but collective efforts to engage with, extend, and enact Bacon’s larger project. Rather than treating the Novum Organum as a self-contained treatise, these readers approached it as a repertoire of experiments (“instances”) and methodological exemplars, and as a “to-be-completed” fragment of the broader, unfinished Instauratio Magna. They became active collaborators, interpreting, testing, and “relieving” (Beale, 1666) parts of Bacon’s work, subsuming the results into their own projects.”

In the discussion of the paper, we considered whether the prompting of such interactive reading might have been a more common strategy. I vividly remembered Locke stating that he makes his argument for the origin of ideas dependent on the readers’ cooperation:

“I suppose what I have said in the foregoing Book will be much more easily admitted, when I have shown whence the understanding may get all the ideas it has; and by what ways and degrees they may come into the mind;—for which I shall appeal to every one’s own observation and experience.” (Locke, Essay II, I, 1)

This appeal is not a mere trope. The argument relies on the readers’ experience. Likewise, Descartes’ Meditations are not a solitary exercise. Apart from the fact that Descartes was adamant that they be published with the Objections and Replies, he insists from the beginning that the reader meditate with him. (See this post and podcast for Andrea Sangiacomo’s take on meditation and the Meditations.)

As Catarina Dutilh Novaes has shown in her The Dialogical Roots of Deduction, logical and mathematical thinking have often been designed as dialogical activities in ancient and medieval contexts. While she paints an intriguing picture of this practice, she repeats the assumption that early modern authors pushed it to the fringes in favour of a mentalistic and individualistic understanding of reasoning (see this paper for a bit more discussion). As I see it, the prompts of collaborative reading show that this picture needs correction.**

At the workshop I asked Dana what she thinks might make us so ignorant of this collaborative strategy in early modern texts. Her reply was that readings inspired by 19^th-century assumptions about philosophy being a systematic endeavour might have contributed strongly to this. Indeed, if you think that a text gives you a system, you will assume that it was designed with the aspiration of providing a systematic whole. Something that does not rely on the (contingent and only partly predictable) cooperation of the reader. In this systematic tradition, it’s the text that provides the argument, and the job of the reader is to read the argument off the text. If this assumption is correct, philosophical texts that come with the ambition of presenting a whole and perhaps closed system will involve fewer or no prompts to the readers (I haven’t checked for this properly, but it’s something that suggests itself).

Of course, this doesn’t mean that the 19^th century leaves us with a final shift to systems or at least to texts that merely count on the readers’ silent comprehension. If you think of texts appealing to future readers, as for instance in Nietzsche or in attempts at ameliorative conceptual engineering, you recognize something like a trend of collaborative reading, named “philosophic prophecy” by Eric Schliesser. According to him, philosophic prophecy is involved in the business of coining concepts that “disclose the near or distant past and create a shared horizon for our philosophical future.” In this spirit, we might think of such texts as prompting collaborative concepts, arguments or entire collaborative projects. Accordingly, it might make sense to distinguish, inter alia, between texts that prompt active reading as opposed to texts that merely demand passive reading (at least for the fulfillment of the invoked argumentative purposes).

Why don’t we recognize this when we’re met with it, then? Does the common paper model, favouring the defence of claims, make us blind to these reading strategies? Be that as it may, perhaps the advancement of prompts for LLMs will raise awareness of this feature again.

______

* I’m enormously grateful to everyone at this fabulous workshop. Special thanks to Dana Jalobeanu and Valentina Sperotto for discussing this mode of early modern reading.

** Catarina Dutilh Novaes makes the following clarification (on facebook): “Just to clarify, my claim about the mentalistic conceptions of reasoning in the early modern period pertain primarily to logic, and how they talked about logic. It doesn’t mean that in their practices of reasoning, i.e. when engaging in philosophical inquiry, these authors were not themselves using dialogical, social strategies.”

Educational aspirations in the digital age

What can we really learn from LLMs? Politeness, I guess

What makes LLMs so successful? I don’t mean their functions in pattern recognition and so on. I mean: why do we or at least many of us use them in so many contexts? I guess they seem to provide something that we often miss in our interactions with fellow humans: politeness. – Let me explain: Whenever I read social media posts, I notice that the tone is often quite problematic: Going by the common modes of expression, we’re often surrounded by dogmatists, cynics, know-it-alls, bullies and other short-tempered fellows. Now, I doubt that people usually behave “in real life” as they tend to do on the internet, but the trouble is: on social media, style is all we have.

My guess is that we’re not exactly made for this disembodied kind of communication: we lack typical cues of body language and the humor of face-to-face interactions or the restraint that we once learned to exercise in mail exchanges. Contradicting your interlocutor with a friendly smile might be fine, as might be breaking troublesome news in a sincere piece of writing. But contradicting a stranger who hasn’t quite made their point yet or bursting out with bad news on the spot doesn’t go down well. Although we do it all the time now, we’ve not adapted well to communicating this way.

By contrast, LLMs are made exactly for this kind of environment. In fact, they have no other environment called “real life”. There is quite a bit of discussion on the politeness of LLMs, but it’s clear that they have been trained to react with polite and partly even deescalating language. Now, while I don’t think that the polite language is the reason we initially turn to these devices, it’s what makes us or some of us stay with them. It’s utterly frightening to see how some people seem to assume that they’re entering real therapeutic or romantic relationships, but given the typical reactions of LLMs the attraction is quite understandable.

Thinking about this on and off, I’m inclined to conclude that these LLMs satisfy a need we have especially in the online culture: politeness. This conclusion came as a real surprise, though. After all, politeness is often mistaken for mere formal behaviour and thus distinguished from real friendliness. In a culture that often prides itself on “authentic” rather than polite communication, I would have thought that chat bots prompt cynical reactions rather than any temptations. But the standards of authenticity don’t work in online interactions with strangers that don’t belong to one’s tribe. There, interactions with people trying to sound authentic prompt more often than not hate or misunderstandings at best. In this environment, the most formal blather of chat bots mimicking politeness seems to be rather attractive.

The moral is: let’s practise politeness – online as much as offline. In online contexts, this means first of all practicing polite conversation. (Having just finished a paper on the notion of conversation, I realise that polite conversation is one of the great achivements of humanity that might be under threat through digital communication.) And one of the first things to remind ourselves of is that conversations do not, at least not first and foremost, consist of claims, arguments or evidence. In this spirit, let me finish with a passage from Michael Oakeshott from his The Voice of Poetry in the Conversation of Mankind (1959):

“We are urged, for example, to regard all utterances as contributions (of different but comparable merit) to an inquiry, or a debate among inquirers, about ourselves and the world we inhabit. But this understanding of human activity and intercourse as an inquiry, while appearing to accommodate a variety of voices, in fact recognizes only one, namely, the voice of argumentative discourse, the voice of ‘science’, and all others are acknowledged merely in respect of their aptitude to imitate this voice. Yet, it may be supposed that the diverse idioms of utterance which make up current human intercourse have some meeting-place and compose a manifold of some sort. And, as I understand it, the image of this meeting-place is not an inquiry or an argument, but a conversation. In a conversation the participants are not engaged in an inquiry or a debate; there is no ‘truth’ to be discovered, no proposition to be proved, no conclusion sought. They are not concerned to inform, to persuade, or to refute one another, and therefore the cogency of their utterances does not depend upon their all speaking in the same idiom; they may differ without disagreeing.”

Friedrich Wilhelm Scharffenberg: Die Kunst des Twitterns (Chemnitz 1723):

How to search effectively for literature when doing philosophical research

I wrote much of my course work on a mechanical typewriter and I learned the ropes of academic literature search by being a student assistant in our faculty library, attributing keywords and preparing card indexes of incoming books and journal issues. What can I say? That world seems to have gone, but the principles still run through our profession. But since googling seemed to have made searches so easy, the art of bibliographing is now hardly ever taught and the word “bibliographing” is listed as rare. In what follows, I want to compensate for the lack of instructions by providing a bit of advice on how to search for primary and secondary literature.

Don’t start too soon! – It might sound odd, but my primary advice is that you shouldn’t start searching for literature too soon. I can never follow this advice, but I actually think that restraining yourself until you’ve worked out a draft proposal for the kind of research you want to do, helps you in sticking to your actual thoughts and doesn’t make you question the legitimacy of your own thoughts too soon. My advice is to wait till you have an actual research question (to be refined later) that actually gives you a clear and sufficiently narrow scope for determining proper keywords to search for.

Searching for primary literature. – Even if you don’t speak the language in question, always take at least a look at the original text in the form of a critical edition. This gives you information on how the text was actually established, whether there are variants or emendations and such like. A text is never a given. And you should be aware that it’s been produced by people with certain aims in mind. The latest critical edition should provide the most faithful text that you can actually quote and reference in your work. So even if you work with some outdated online version, make sure to reference the work in question in keeping with the latest critical edition.

How do you find a proper critical edition? – One way is to do an online search and look at scholarly pages. It might be the Stanford Encyclopedia of Philosophy (SEP) or some historian of philosophy working on the author. Begin by checking the pertinent bibliographies. If that doesn’t yield results, check for recent secondary literature and see what’s being quoted. (Note that the SEP, while a great tool, often merely comprises anglophone literature, shunning entire research traditions in other languages.)

Translations. – When using translations into your native language, make sure to check several ones. Unlike critical editions, translations often come with very different aims. First of all, be aware that a translation is to some extent the production of a new text that might – for various reasons – be more or less close to the original. Most importantly, translations establish their own terminological traditions that are handed down in some quarters but not in others. Finding your way requires you to check what the translator says about the rationale behind the translation in question.

Searching for secondary literature. – Besides using the proper critical edition it’s the command of the relevant secondary literature that can make you stand out as a diligent reader and scholar. Honestly, checking the bibliography of a piece tells me quickly whether someone’s done the required work. Anyway, the first step for you should be narrowing down a list of keywords. That’s why a proper research proposal is crucial. Make sure to bring down your list of keywords to the actual aspects you’re working on:

A straightforward approach is this: Insert all the keywords in your research question or thesis.
You might even put in the title of your own work in the search engine. So if you search for something on Locke on language, check “Locke on language” (with quotation marks!) first.
Then try those keywords in different languages (you master).
Then try inserting key phrases or passages from sections in the texts you’re working on (again, with quotation marks!). This way, you’ll find works that discuss the same stuff as you. Try the original and translations.
Also think of cognates and related terms to the keywords you’re searching. Often, it might make a difference whether you’re searching for “sentence” or “proposition”.
Try also inserting pertinent scholars or related authors with the pertinent keywords.

Obviously, there are many more ways. What’s important is to keep a balance between systematic keyword-related and spontaneous association-related searches. Often, crucial impulses result from random results.

Kinds of sources. – What you’ll find this way, will vastly differ: from scholarly lit erature to blog posts to videos etc. When exploring, anything might be helpful. When writing up your piece, make sure to assess whether what you’ve found is up to standard. Scholarly articles are fine. Often, blog posts and videos might be equally fine. To refine your own bibliography, make sure to check the bibliographies of the works you’ve found. And don’t forget that there are even professional and published bibliographies. In any case, when searching with AI tools, always make sure that the actual source exists.

Engines. – Obviously, there are different search engines (even on google, google books, google scholar etc. will yield different results) and databases of libraries, of research institutions, and even of individual scholars. Make sure to figure out patterns of results you get with those and use them accordingly.

Serendipity in libraries. – Finally, don’t forget that the crucial thought that might help you might be in a place you’re not actually searching. When I started my searches in the library back in the day, I often realised that some crucial idea was not in the book I had singled out from the shelf but in the book right beside it. This kind of serendipity sometimes got me further than a methodical search.

Ultimately, the point is that literature search is something worth of attention in its own right. Here, I’ve only scratched the surface. If you have further tipps, please put them in the comments.