Dot Objects: Don’t Judge a [____] by its [____]

There’s an obscure concept (from an obscure field called semantics) that I find really fun to think with: Dot Objects. This post is an attempt to pull it out of that technical field and into, well, the community of people who read my blog. I think that semantics tools are fundamental for rationality and quality thinking in general—Alfred Korzybski, coiner of the phrase “the map is not the territory” and founder of the field general semantics, would probably agree with me. Note that I extrapolate a ton here, so (disclaimer!) don’t take anything I say as being true to the technical study of the subject.

So. Consider the sentence: “The university needed renovations, so it emailed its alumni to raise funds.” The university that has the alumni isn’t the one that needs the repairs. One is an organization, the other is a physical structure.

Dot objects are

entities that subsist simultaneously in multiple semantic domains.[1]

The name “dot objects” (also sometimes “dot types”) comes from the notation used in academic papers on the subject, which is X • Y where X and Y are the two domains. So the above example might be OrgPhy.

Aside: this is the third post in a month or so that has contained Xs and Ys. Seems like decent evidence that my blog is getting more abstract.

Interesting Examples

Meals exist both in the event domain and in the food domain. So you can say something like “lunch went way too long but was delicious”. Meals might also exist in a social or information domain. Like the “lunch” can also be “interesting” or “worthwhile”. These could refer to the food, but more likely to the conversation. And of course they have a physical dimension, which can be “beautiful”. (EventFoodPhy)

Cannabis exists in the physical domain (a specific instance of the plant, or perhaps its physical shape in abstract), in the biological domain (the sense in which one can say that it is a relative of the hops that are in some beers) as well as in the domain of psychoactive compounds. Similarly, if we were to say “a hot cup of coffee woke Frida up” then we’re guessing Frida drank the coffee (Food) but that it was the caffeine contained in it (Drug) that woke her. Unless it’s decaf coffee, and the effect is placebo. Or if the coffee was poured on her sleeping body, which would also be a reasonable (just less likely) interpretation of that sentence.q

The “cup”, incidentally, in the earlier sentence, is also kind of a dot object. “Cup” refers to both the contents and the container. Another example: “I walked through the door then closed it.”

Analogously to a dot object, we might talk about dot actions, which are actions that can be understood in multiple domains. A handshake, for example, is a physical gesture, but it also has a social meaning (PhySoc). Writing is both a physical action as well as an action in the domain of information (PhyInfo). I started drafting this blog post a long time ago, and can’t recall if I read about dot actions somewhere or just made this up.

Meta-semantics: applying this to words

Cognitive Lode, a fantastic website highlighting the latest results in cognitive science and psychology (caution: therefore the ones most likely to be due to statistical error or similar, though this one has a huge effect size so maybe it’s okay) recently posted about what they call the Bye-Now Effect:

Research just released suggests that when we read words like ‘bye’ or ‘wait’, we automatically think of and act on words that have the same sound, such as ‘buy’ or ‘weight’, especially when we’re tired.

This can be modelled using dot-objects. It’s a bit of a stretch, but if you consider “buy”, it exists in several domains:

  1. Strings of characters: the three letters it contains, arranged in that order
  2. Sounds: /baɪ/
  3. Meanings: “action of purchasing something” (and other meanings such as bribing or accepting, eg. “yeah… I don’t buy it”)

The effect occurs when “bye”, which shares reference #2 with “buy”, gets semantically converted into “buy”, and gains reference #3 as a result. Upon reflection, I guess this just means that homophones are sort of a special case of dot-objects, which is actually not that exciting. Still, a neat abstraction.

Don’t Judge a Book by its Cover

Consider the phrase “don’t judge a book by its cover”. It’s essentially saying “don’t judge 〈book, Info〉 by 〈book, Phy〉”. As a maxim, it is applied metaphorically to mean not judging the contents of something from its external appearance. Can we get anything new out of it if we port it to other domains? We might end up with maxims like:

  • “don’t judge the quality of someone’s character by how they dress” …about the same as the original
  • “don’t judge a… durian by its smell?” …durians are a stinky fruit. Personally, I don’t like the taste either. This is also a bad heuristic for most food.
  • “don’t judge the trueness of a religion by the opulence of its cathedrals” …this is slightly different, I think. I happened to visit a Basilica once, and it was tempting to say “clearly there must be a [Catholic] god for this building to be so amazing.”
  • “don’t judge someone’s trustworthiness by the firmness of their handshake” …also distinct from the original, though barely.

It happens, of course, that in many cases there is some correlation between these two factors, such that one of them is (perhaps weak) evidence for the other. But, as with books, it’s important to recognize that they aren’t the same thing, even though its easy to conflate them. unconsciously.

The Self

The self is the dottiest object I’m aware of. It exists in so many domains. There is the physical self: the matter that composes one’s body (“I poked Bob”). Similarly but distinctly, there is the genetic self: the DNA of an individual (“Scientists cloned Jane”). There is the self of personality: what we might mean if we said “a Malcolm-bot”. Then, for most people, there is at least one self of identity. And maybe a self, beyond identity, that observes the identity.

When I was 15 I read a book by Eckhart Tolle, that had a huge impact on me. I’m no longer particularly connected to its spiritual aspects as such, but I still remember a story Tolle tells, where he does a really cool metacognitive thing (which prompts his enlightenment or whatever). He described himself feeling deeply depressed and suicidal, and one thought kept playing on repeat: “I can’t live with myself.” But then he concluded that in order for that sentence to work, there had to be two of him. And he suddenly blurted out,

“If that’s the case, then who is the ‘self’ that ‘I’ cannot live with? And who am ‘I’?”

Consider also the “self” that is Bill Gates. Bill Gates has all of the aforementioned self-aspects. But “Bill Gates” also refers to a concept that many people have in their mind, of the richest person in the world. Facebook distinguishes this using Pages versus Profiles. There is a Bill Gates “public figure” page, with 11 million Likes. It’s reasonably likely that he also has a personal profile, that he uses to connect with his friends. This is the same person, and yet… it’s not. The brand of Bill Gates exists in a totally different semantic domain than the friend Bill Gates.

An animated gif of a comedy stage with Mark Zuckerberg on the left, Jesse Eisenberg in the middle, and Andy Samberg in the middle. Samberg is mouthing 'awkberg' while that word flashes in the front in big text.

While we’re talking about rich tech company cofounders and about Facebook, consider Mark Zuckerberg. He has all of the things we’ve described so far. He also has a movie about him! So there exists a “Mark Zuckerberg” the character, which in the feature film was played by Jesse Eisenberg. Zuckerberg has also been uhh, played by comedian Andy Samberg, who impersonated him for one of his keynote talks. (Incidentally, they all found their way onstage at the same time once, resulting in the gif on the right). Anyway, one can’t “play” a person in the other senses, so there’s an additional domain here of character.

And, related to brand, though distinct, we could talk about the philosophical domain of someone. For the most part, when people talk about Ayn Rand these days, they’re not talking about the person-who-lived so much as they’re talking about Objectivism, the philosophy she’s associated with.

So far we’ve got something like Self: PhyGeneIdEgoObsBrandCharPhilosophy

Anything else? Suggest more in the comments.

Technical sources on this subject, probably not worth reading unless you’re big into language processing or linguistics:

If you found this thought-provoking, I invite you to subscribe:    
About Malcolm

Constantly consciously expanding the boundaries of thoughtspace and actionspace. Creator of Complice, a system for achieving your important goals.

Personal Website


Staale Nataas » 5 Dec 2014 » Reply

Interesting article. The “bye-now” effect that you describe has been taught and used in NLP for decades, mainly for hypnotic induction and persuasion. An example of the former might be a sentence like “You’re unaware of you’re unconscious now”, the last three words constituting a command masked by a seemingly innocent sentence. A example of the latter might be emphasizing certain words in a separate tone like “This guy who passed *by me* had a strong *presence*”. ( Buy me presents).

Nancy Lebovitz » 3 Jan 2015 » Reply

Thomas Hanna distinguishes between the body (physical object) and soma (body as experienced by a person).

    Malcolm » 3 Jan 2015 » Reply

    Ooh, I really like that. So would you talk about having a “sore soma”? Or am I totally butchering it? 😛

Have your say!

Have your say!


Name *

Email *