A significant component of human intelligence is our ability to use symbols. In this article, I would like to discuss a few things about symbols that I think can sharpen our ability to reason and communicate if we keep them in mind. Most of these things may be common knowledge, but we can sometimes gain new insights just by considering a well-known idea in the right context so I would like to explicitly record them here.

What is a symbol?

A symbol is anything with which we mentally associate another object or concept. It can take many forms, including spoken words, written words, images, and gestures. Symbols allow us to communicate; we cause others to think of something by producing the symbol that is associated with it. When a person hears or sees or otherwise observes a symbol, they tend to think of a particular concept. Seeing an object can likewise cause you to think of the symbol associated with it.

This reaction, thinking of a concept when you see a symbol, is a physical process that takes place in our brain and nervous system as a whole. Being a physical process mean that it is, to a large degree, deterministic. It also means that it is dependent on the state of the brain; the particular reaction that the brain has to a symbol owes to the physical configuration of the brain’s parts.

The definition of symbols is arbitrary. “Correctness” of a definition only means whether it conforms to the definition that someone else chose.

How definitions are created

Recalling that a definition is the effect of a particular physical configuration in the brain, it follows that a definition is created when that configuration is formed. Our brains do this automatically; when we see things together frequently, our brain will naturally form connections that cause us to be more likely to think of one thing when we see the other.

I think it is useful to consider two scenarios in which symbols are acquired; that in which the definition’s object is present, and the other in which it is not.

Object is Present

We acquire many of our first symbols in the presence of the object. For instance, a parent shows us an apple and says “apple”. Though obvious, it is worth mentioning that when a symbol is created this way, then it stands for a real object.

Object is Absent

In order to define symbols in the absence of their object, we must use other symbols. This of course requires that we have already learned those symbols. In contrast to the previous scenario, the symbol’s intended object may not exist, even if those of its constituent symbols do. For example, horses exist and horns (such as that on a rhinoceros) exist, but a unicorn, defined as a horned horse, does not exist.

In the absense of an object, symbols are defined by combining symbols, and consequently concepts. Each additional symbol or concept can potentially narrow the set of objects that are described by that combination, or even empty it completely as in the case of the unicorn.

Object is Simulated

For completeness, I’ll discuss a third scenario where the image being defined is absent but simulated in some way. This scenario is, strictly speaking, a subset of the previous, but it has characteristics not shared by the the rest of the ‘absent’ scenario. By ‘simulated’, I mean that an attempt is made to accurately reproduce some sensory pattern that the object itself would. Examples of this are photographs, drawings, sound recordings, or an imitation of a sound the object makes. These can be very effective ways of defining something so that a person will be able to recognize and identify the actual object when they encounter it.

Chosen and Consequential Characteristics

Chosen Characteristics

I think it is essential in communication to discern between statements that convey a symbol’s chosen characteristics from those that convey its consequential characteristics.

Chosen characteristics are (as you would expect) those that a symbol’s objects possess because we decided they do. It might be more accurate to say that we decided to associate the symbol only with objects that possess these characteristics. Again, it is possible that there are no objects with a given set of characteristics, in which case we would never find an object that we can correctly associate the symbol with.

[Talk about chosen characteristics of “Object Present” definitions]

Consequential Characteristics

Sometimes there are objects that possess the characteristics we chose for a symbol. In that case, those objects will often have other characteristics, outside the chosen ones. These are what I’m calling consequential characteristics. We didn’t intentionally choose to associate the symbol with objects having these characteristics, it’s just a consequence of our choice. For example, apples have cells, but people have had a name for them since before we knew about cells, so that particular characteristic is not chosen, but a consequential.


Statements expressing the chosen characteristics defining a symbol do not convey information about its objects, as those objects might not even exist. Such statements instead convey information about us; they convey that we have made an association which, again, is a physcal configuration of our brain.

It follows that to convey information, we must make statements about the consequential characteristics of objects associated with our symbol. This statement can be as basic as saying that those objects exist.

It is also meaningless to ask whether the objects associated a symbol have a given characteristic when that symbol has no chosen characteristics, since without them, the symbol doesn’t refer to anything (or maybe it refers to anything, which would be equally useless). To understand this, imagine that someone walks up to you and, without any context, asks “hey, was that thing green?”. You don’t know what that thing is, so you can’t possibly know whether it was green. Yet, since chosen characteristics are what specify a symbols object, using a symbol without chosen characteristics is exactly like this.

This may seem abstract and unimportant, but I have seen many discussions carried out where a central term was not defined, or where the participants apparently had different definitions for it. In many situations, if this problem were resolved, there would be no disagreement at all. It is as if two people were having a disagreement about what a painting looked like, while they were not looking at the same painting.

I will be referencing these ideas about symbols in future posts.