Language Games and Particles

There’s a phrase I hear every now and then that always makes me curious about how English will look in the future. The phrase goes like this, “So are you going to come here, or…”, where the part before ‘or’ can be substituted with any yes/no question. It is always uttered by someone who has a vested interest in the question being answered affirmatively.

It’s interesting because it’s such a nice, self-contained example of the kind of linguistic metagames (more on this later) that we’re constantly playing. There was a kind of predictability about the situations in which this phrase would make an appearance, and I thought I would try to understand what was hiding underneath.


The crucial part of this phrase is that the sentence is never completed, and the speaker never specifies what the possible alternative after the ‘or’ could be, yet everyone involved in the conversation has at least some idea of potential alternatives, and is acutely aware that the speaker desires none of them. In the absence of any words to communicate a judgment about the situation, the form of the utterance and the fact of its incompleteness do the work instead.

This kind of mechanism is similar to how grammatical particles work in lots of other langauges. One common example is the interrogative particle 吗 (ma) in Mandarin Chinese, which is tacked onto the end of a sentence to turn it into a question.

你   喜欢   吃   苹 果   吗?

you  like  eat  apples  interrogative

‘Do you like to eat apples?’

Without the final character representing the interrogative particle, the above clause would be a statement that could be translated as “You like to eat apples”. The way the 吗 particle works is also roughly equivalent to the English ‘huh’, as in, “You think you have what it takes, huh?” although arguably the inflection of one’s voice does a lot of the heavy lifting in that example.

In the case of 吗, it’s a ‘word’ that doesn’t have any meaning by itself, but affects the semantics of the surrounding sentence. In English there are also other types of particles, for example those used to punctuate our sentences as in ‘So’, ‘Well’, ‘Right’, ‘Oh’, etc.

I see ‘or’ eventually becoming something similar to ‘well’, which has a relatively clear set of meanings (i.e. a noun denoting a hole in the ground, an adverb meaning ‘done in a good way’, etc.), but is also often used to communicate a kind of judgment. Here I’m referring to the fact that starting a sentence with ‘well’, can connote disagreement or hestitation to agree with one’s intelocutor (among many, many other things), as in:

A: You need to find a job.

B: Well, I have a huge following on Instagram.


This kind of transition has already taken place with the word, ‘please’, although we rarely think about it. The full, unwrapped version of ‘please’ used to be, ‘if you please’, or ‘if it pleases you to oblige me, …‘.

The French, “S’il vous plait” has preserved this construction exactly. English, despite having the same once upon a time, has truncated it but kept its meaning; the shortened ‘please’ has now become something greater than a syntactic shortcut to the original, since we now recognize ‘please’ itself as a politeness marker, without ever thinking of its historical form or meaning.

John McWhorter has an interesting TED talk about this phenomenon, where he discusses the evolution of language, specifically the emergence of new particles like ‘lol’, as a consequence of SMS messaging.

Politeness Theory

So, coming back to ‘or’, how can we understand what is actually happening here? There’s a framework that can be used to model this type of construction which comes out of Politeness Theory as proposed by Penelope Brown and Stephen C. Levinson in 1987.

Since it’s been around for a while, plenty of modifications to the original formulation have been proposed in order to account for how politeness works in other languages besides English, but the original concepts can be applied very effectively to the construction in question.

At the center of politeness theory lies the concept of ‘face’, like the kind that you can save or lose. Face comes in two flavours, positive and negative, with positive face being “the want of every member that his wants be desirable to at least some others executors” and negative face being “the want of every ‘competent adult member’ that his actions be unimpeded by others”. This can be simplified by saying that positive face involves everyone’s desire to be liked by others, and negative face involves everyone’s desire to do just what they want.

Every utterance that an individual can produce carries within it the potential to threaten someone’s face, positive or negative, hearer or speaker. The term for this potential to damage face is, ‘Face-Threatening Act’ (FTA). The puns afforded by this concept are really too deliciously irresistible to pass up :)

FTAs are pretty straightforward - if I ask you to do something, then your negative face is threatened (now you might have to do something you don’t want to do), and, of course, depending on our relationship and the nature of the request, it might threaten your positive face (your desire to be regarded positively) if I’m asking you to do something that you consider to be beneath you.

On the other hand, if you refuse it could also threaten both of our faces. For instance, my positive face is threatened (clearly you don’t think very highly of me if you won’t do what I ask), and potentially your negative face is threatened (be refusing you might now be in my debt, and have to find a way to make it up to me).

Faced (whoops) with the possibility of committing an FTA, we have to find a way to mitigate it, which might lead to an exchange like this:

A: Hey, I’m really sorry to bother you, but I’m going away for the weekend, would you mind feeding my gerbil while I’m gone?

B: Oh gosh, you know, I’d love to, but I’m allergic to gerbils, sorry, I’d love to help, but no can do.

Apologies come from both sides as both sides have inconvenienced each other and want to avoid paying the social penalty for doing so - both sides want to have positive status in the eyes of the other.

Now let’s model it using the ‘or’ construction.

A: Hey, I’m really sorry to bother you, but I have a favour to ask.

B: Sure, what’s up?

A: Well, I’m going away for the weekend, and I have a gerbil, do you think you’d be able to feed him while I’m gone, or…

B: Oh gosh, you know, I’d love to, but I’m allergic to gerbils, sorry, I’d love to help, but no can do.

Or what? Or do I have to make alternative arrangements because you refused, which is profoundly disappointing, and frankly as far as I’m concerned, your (hearer) positive face has crumbled into dust?

By committing this FTA, the speaker is leveraging the hearer’s desire to maintain positive face to persuade the hearer to indulge the speaker’s request.

At the same time, explicitly spelling out what comes after ‘or’ would be a very hostile act indeed, and given the roughly equal relationship between the two participants in the example, is something that would cause significant harm to the speaker’s positive face.

This is what deploying the ‘or…’ is really trying to avoid. And the way it’s avoided is with plausible deniability - you think you know what I was going to say after “or”, but you can’t prove it, because I didn’t say anything at all and just let your imagination fill in the gap.

The Meta

How cool would it be if all of this nuance could be encapsulated in a discrete pragmatic particle? In addition to the grammatical meaning of ‘or’, there could be this whole other meaning that you get for free as soon as you tack it on at the end of something question-like. In its eventual form, it would become for persuasion what ‘please’ is for politeness - a stand-alone unit that now represents politeness by its very existence, not by its being a shorthand for some other, more ‘genuinely’ polite construction.

One way that language changes over time is that some meanings shift from initially finding expression as a consequence of a subtle arrangement of grammar, vocabulary, and the surrounding context(s), to coalescing into a definite, concrete form. Instead of a diaphanous film hovering above the surface of literal meaning, this semantic content can come to be explicitly captured in a word or particle.

This is one of the most fascinating things about language, it is so lusciously context-dependent - there are layers upon layers of meaning, and since some information is too delicate to be conveyed literally, we must continually find novel ways of making it tacitly known, as our vehicles of non-literal expression become compromised.

This is the metagame of language.

Cedric Chin wrote a fantastic article about the concept of metagames or ‘the meta’, and how they exist in everything from card games to business. The article is great, and you should absolutely read the whole thing. The core idea is that mastering any domain requires mastering not only the basic operations permitted in the domain, but becoming adept at handling the changes in what operations are permitted/effective and the resulting downstream effects for the effectiveness of all other basic operations.

What I’m really curious about is how certain ways of conveying information transition to and from the status of a basic operation - in this case invoking a literal meaning of a word (and we can disregard for the time being how inconceivably far from basic this is) - and the status of high-level strategy (the metagame), which in this case would be the arrangement of particular words in a particular sequence, given a particular pragmatic context, until a deliberate meaning emerges.

There’re myriad parts of our language undergoing these transitions, spurred on by the habits of one group of speakers or another. I could easily see ‘or’ undergoing a transition like that.

Or perhaps it could begin with the similar usage of ‘if’. I’ve also encountered an interesting construction that goes something like, “So if you could do that”, stated as a complete thought with nothing following after.

But who really knows what perfectly ordinary part(s) of our current language will be transformed into something unrecognizable by future speakers. The transitions will seem gradual to us, but perhaps in retrospect the contrast in our usage will be unmistakeable.

At the very least, if we needed another reason to listen actively to others, it can be to attend to how they wield their language, and wonder at the miracle of the mind that makes it possible.


Chin, Cedric. To Get Good, Go After The Metagame. 26 Feb. 2020,

McWhorter, John. “Txtng Is Killing Language. JK!!!” TED, TED, Feb. 2013,

Meyerhoff, Miriam. Introducing Sociolinguistics. Routledge, 2011.

© 2024. Ilya Meerovich