In the 1930s, the Polish engineer, Alfred Korzybski, developed a theory of abstraction and evaluation called General Semantics. It included training methods designed to clarify thinking in a way that was believed to lead to more adaptive emotional responses. Korzybski’s work was a big influence on Albert Ellis, who later developed what became known as Rational-Emotive Behaviour Therapy (REBT), the original form of Cognitive-Behavioural Therapy (CBT), in the 1950s.

Back in the heydey of psychoanalysis, Korzybski developed a quasi-therapeutic method, a general form of self-improvement training, that was poles apart from Freud’s approach to psychotherapy. General Semantics approaches self-improvement, and interprets psychotherapy, from the perspective of mathematics, logic, and scientific method. It can be viewed as one of the main precursors of modern CBT, which has validated some of its key concepts, although, in my experience, General Semantics still offers some useful techniques.

In this article, rather than discussing General Semantics in depth, I’m going to give some examples of ways in which it can be helpful when dealing with emotions.

Extensional Language

Korzybski’s most famous saying was “The map is not the territory”, which means words are not the events they represent. He believed that the human organism is designed to abstract information from its environment selectively. That allows us to process information and communicate more efficiently but it also means that we’re prone to confuse levels of abstraction. In particular, when we identify abstract thoughts with real events, fusing them together, we become susceptible to emotional problems.

In logic and philosophy of language, the “intension” of a word or phrase signifies its meaning as expressed by other words. For instance, the intensional meaning of “dog” is “a four-legged domesticated mammal of the genus Canis lupus familiaris.” The extensional meaning of the term, by contrast, would be the actual things to which it refers — something you could indicate by pointing at one or more real dogs. Korzybski introduced several techniques for “extensionalizing” our thinking, describing events more accurately, in a way that’s grounded in non-verbal experience, or what he calls the “silent” level of experience. This requires separating our inferences and evaluations from observable objects and events.

Let’s start with a simple example. Suppose that I become angry with someone because they don’t agree with what I’m saying. Initially, I may say to myself,:

This guy made me angry, he’s a total idiot!

Focusing on that way of looking at the event appears likely to make me continue feeling angry, and it may lead to other problems as well, causing me to get stuck in a loop.

Suppose, however, that I recall the event and separate my thoughts from the observable event by placing them in quotation marks.

I heard that guy disagree with what I said. I then added the inference “He doesn’t respect me”, and the evaluation “That’s AWFUL, he’s a total IDIOT.”

Take a moment to pause over that description of the event. Already, it takes slightly longer to process. Korzybski believed that extensional language introduces a delay that supports more reflective evaluation.

That this [neurological] ‘delayed action’ is beneficial is acknowledged by the majority of normally developed adults in the form of [observable] delay in action and finds its expression in such statements as ‘think twice’, ‘keep your head’, ‘hold your horses’, ‘keep cool’, ‘steady’, ‘wait a minute’, and such functional recipes as ‘when angry, count ten’. — Korzybski, Science and Sanity

Delay, in other words, is only of value insofar as it allows us to re-appraise our semantic evaluation of an event. Korzybski hypothesized that this required better functional integration between cortical systems involved in higher-order discrimination, delayed response, and reflective problem-solving, and subcortical affective systems associated with faster, more automatic emotional reactions. He described this crudely in terms of the “cortex” and the “thalamus,” drawing on terminology from the basic neuroscience available in the 1930s, whereas modern psychologists would frame things somewhat differently. Although his terminology is dated, the point he’s making is still very relevant to cognitive therapy.

Sometimes this is referred to as the “Semantic delay”, in which we slow down, by using extensional language, and buy ourselves time to become more aware of the distinction between our thoughts and real events. Korzybski called this “consciousness of abstracting” and General Semantics can be viewed, to a large extent, as a method of training in consciousness of abstracting by using language with greater precision and self-awareness.

In addition to using quotation marks, Korzybski also recommended using the phrase etc. or et cetera as a symbol of “non-allness”. It reminds us that our thinking, by its very nature, involves abstraction and is highly selective.

I heard that guy disagree with what I said. I then added the inference “He doesn’t respect me”, and the evaluation “That’s AWFUL, he’s a total IDIOT.” et cetera

In other words, there’s always more to reality than our thoughts can capture. All thinking is selective thinking. You could also say “… and there’s more to the story.” That’s true even at the level of raw sense experience. Our senses only reveal part of what’s going on around us. Other people might see, hear, and otherwise perceive things differently than we do or notice things we missed. We only experience a miniscule fraction of the wider context, which potentially changes the meaning of events. As we abstract further, and make inferences and evaluations, we leave out more and more information, and our thinking becomes even more selective.

It may seem pedantic at first to spell this out but think of the effort required as a means of engaging different regions and systems of your brain. When we remain conscious of the selective nature of our thoughts, we potentially weaken the identification between words and experience, and allow our emotional responses to become more flexible and adaptive.

Korzybski also suggested using dates to mark events, objects, and people more precisely, in order to counteract our tendency to fuse different moments in time together. General Semantics viewed people, objects, and events as dynamic processes unfolding over time, and dating is therefore intended to increase our consciousness of the distinction between different moments in those processes.

Last Tuesday, at 3pm, in the office, I(27/12/2025) heard Tom(27/12/2025) say he disagreed with what I said. I then added the inference “He doesn’t respect me”, and the evaluation “That’s AWFUL, he’s a total IDIOT.” etc.

Again, this may seem laborious, but in a sense that’s the point, or at least it’s part of the procedure. It takes a few seconds longer, and requires marginally more effort, but also encourages the brain to respond less hastily and less emotionally, by involving more of the neocortex in the experience. Korzybski was inspired by logic, mathematics, and the scientific method, in this regard. His major work is, in fact, called Science and Sanity (1933) because he believed that a more scientific way of thinking and using language could potentially improve our mental health.

Pause for a moment to compare the “feel” of the previous (extensionalized) statements with our original statement. Imagine yourself reliving a similar memory, for instance, and telling yourself:

This guy made me angry, he’s a total idiot!

It can help to “shuttle” back and forth between the original way of thinking and the extensional description, in order to study the contrast between the two different mental states associated with these ways of thinking.

Emphasizing the subjective nature of the experience and framing it as relative to a specific time and place can be done in other ways. The General Semanticist, Wendell Johnson, would sometimes use phrases such as “it seemed to me” or “it seemed to Tom”, etc. For example:

Last Tuesday, at 3pm, in the office, I heard Tom say he disagreed with what I said. I had the impression, at that time, in that context, that he doesn’t respect me, and it also seemed to me that it was AWFUL and that he’s a total IDIOT. …but that’s not the whole story.

We can also refer to our internal physiological reaction in the following way, which highlights awareness of the distinction between the bodily (interoceptive) sensations of nervous arousal and the way we interpret them, and label them, as specific emotions.

Last Tuesday, at 3pm, in the office, I heard Tom say he disagreed with what I said. I had the impression, at that time, in that context, that he doesn’t respect me, and it also seemed to me that it was AWFUL and that he’s a total IDIOT. I felt the sensation of my heart beating faster, and tension in my neck and shoulders, which I labelled as “anger”. …but that’s not the whole story.

Labelling our internal reactions in this more granular way can help healthy emotional processing to take place.

The Scary-Looking Diagram

Korzybski created a visual model called the “structural differential” to illustrate the process of abstraction. He believed it was most effective when he physically pointed to the different levels in order to train others in consciousness of abstracting. It’s actually not as scary as it looks. I’ll just provide a very simple explanation below.

The shape (a parabola) at the top basically represents the Event (E) in the real world, before we even perceive it — it’s broken off because it continues off the paper indefinitely. The holes in various shapes represent information and the strings show how information is partially carried through to different levels of abstract thought. The circle near the top represents an Object (O) that we actually perceive through our senses. The luggage tags are verbal descriptions, starting with an attempted Description (D) of the Event, such as “Tom said he disagreed with me”. There are then a series of numerous Inferences (I), which go beyond the observable facts and become progressively more abstract — the last one breaks off because the sequence of abstractions continues off the paper indefinitely.

Finally, the long arrow represents the process of Abstraction (A) being identified with the Event, which veers aside because “The map is not the territory” and our Inferences can never perfectly represent the real Event. If you take nothing else away from this odd-looking diagram, at least it can be helpful as a way to remind yourself to distinguish between real Events and the numerous Inferences (luggage tags) we make about them because our words are not the things they claim to represent. From this perspective, whatever we say a thing is, it is not, because our statements are verbal abstractions from non-verbal events. Reality is not exhausted by language, and to identify words, descriptions, or evaluations with the events they refer to entails confusing levels of abstraction.

Some Links with REBT

Albert Ellis, the founder of REBT, had read Korzybski and the words of later General Semanticists, and was influenced by them. In this article, I’m just going to make one simple observation about the relationship between REBT and General Semantics, though, in relation to Ellis’ famous ABC model: Activation, Beliefs, Consequences. Here’s an example of the incident above in ABC format:

• A: Tom said he disagreed with what I said.

• B: It is AWFUL when people disagree with me, and they’re total IDIOTS for doing so.

• C: Anger.

Already, the ABC model distinguishes clearly between the observable event that activates the response, and the belief that I use to evaluate it. These two things are normally fused together initially so sharpening the distinction between them can be viewed as training in consciousness of abstracting. We could further extensionalize the language, though, by adding the following phrases:

• A: Last Tuesday, at 3pm, in the office, I(27/12/2025) heard Tom(27/12/2025) say he disagreed with what I said. I added the inference “He doesn’t respect me”.

• B: I then added the evaluation: “It is AWFUL when people disagree with me, and they’re total IDIOTS for doing so.”

• C: I then felt the sensation of my heart beating faster, and tension in my neck and shoulders, which I labelled as “anger”. etc.

General Semantics would add awareness that the “activating event” itself, even if described in objective language, denotes something abstract and highly selective. No verbal description of events can ever be complete. We’re leaving out a lot of potentially relevant information about the context in which this event occurred. Training yourself to remain aware of the non-allness even when you describe the “facts” can help to improve cognitive flexibility. It can also be useful to remind yourself that other people may have seen or heard slightly different things — disputing the “universality” of our perspective. We’re not all-knowing (omniscient) beings, so we’re never dealing with the whole picture.

Korzybski would also highlight the problem caused by the “is” of identity — technically the verb “to be”. For example “It is AWFUL”, “They are IDIOTS”, “I am ANGRY”, and so on. This encourages identification, or fusion, between the event and the language being used. Thinking “They are IDIOTS” differs from thinking “I chose to label, or classify, them as ‘IDIOTS’”. The former implies there’s no other conceivable way of viewing them, and thereby traps us in our rigid evaluation, and the extreme emotional response associated with that way of thinking.

Conclusion

In CBT, we often find a simplified version of this advice, such as the practice of describing events in neutral language, without strong value judgements, or to use objective language and stick to the facts. We can also find some parallels between Stoic philosophy and General Semantics. The goal of life for Stoics was “living in agreement with nature”. This phrase literally means “saying the same thing” as nature, in Greek. The Stoics modelled the proto-scientific way of describing events typical of early presocratic philosophers of nature. They describe the goal of their epistemology as phantasia kataleptikai or “objective impressions”, by which they meant descriptions of events that firmly grasp reality.

Epictetus refers to this as “the correct use of impressions”. When he says that it is not things that upset us but rather our opinions about them, he urges us to distinguish more clearly between events and our evaluations of them. Moreover, when someone’s behaviour offends us we should tell ourselves: “He had the impression he was doing what is right” or “It seemed right to him.” Epictetus advises us to speak to our own troubling impressions and say “You are just an impression and not the whole thing that you claim to represent.” We should stick to the facts, that is, and remind ourselves that our impressions are not the whole story.