Can we have an axiomatic theory of psychology?
A bit of background: since at least the time of Spinoza, up until the present day, a limited number of people have made attempts at creating an axiomatic approach to understanding and studying the human psyche.
They have all failed to gain any traction or respect. The most likely reason, in my assessment, is quite simply having way too many axioms.
An axiom is supposed to be a clear, “self-evident” truth, from which one can then logically derive other truths. The most famous use of axioms is by Euclid, where he used 5 very simply stated axioms (like “The whole is greater than the part”) to derive a few hundred pages of geometry. Modern mathematical economics also uses a handful of axioms, while the Austrian economist Ludwig von Mises only required one axiom (“Human action is purposeful behavior”) to understand the world.
However, modern attempts at axiomatizing psychology include dozens of axioms. For example, the “pyscho-logic” approach has over 50 axioms. What’s worse, is that many of these axioms have been expressed mathematically.
To me, this is unnecessarily complicated analysis. So below, I have attempted to give a start to a new kind of axiomatic psychology, using only one axiom:
Axiom: Thinking is the ability of a brain to observe, analyze, and decide.
And now some clarifying definitions:
- To observe is to collect data.
- To analyze is to connect different observations together.
- To decide is to react to the analysis.
We can now derive some truths (a.k.a. theorems) about psychology using only our one axiom (and 4 definitions):
Theorem 1: Thinking takes place over time.
Proof: From our axiom, we know thinking requires observation, analysis, and decision. From our definitions, we know that decisions require analysis, which requires observations in turn. This ordinal process of thinking must happen in sequence; and if it must happen in sequence it cannot happen simultaneously. If a process cannot happen simultaneously, then it cannot take place instantly. Therefore, thinking must take place over time. QED.
Theorem 2: Thinking can be either conscious or unconscious.
Proof: Define unconscious thought as thinking that is entirely determined by biochemical reactions (in other words, the process is entirely automatic), and conscious thought as thinking that is at least partly affected by a non-biochemically determined choice of what to observe, analyze, or decide.
Since these two definitions do not contradict each other, and since they are each and together entirely consistent with both the axiom and all of the other definitions, then we have no reason to reject them as false. And since they are together mutually exclusive and are collectively exhaustive subsets of the axiom, we can therefore make the universal claim that all thinking is either conscious or unconscious. QED.
(Note: we have only argued that it is possible to distinguish between conscious and unconscious thought. Whether a particular thought is actually conscious or unconscious is a different question.)
Now here’s a slightly more difficult one:
Theorem 3: conscious thought requires the faculty of language.
Proof: From our axiom, we know that analysis requires connecting different observations together, and decisions require connecting analysis with behaviour. But to choose what to analyze among a set of different observations, and to choose what action to do among a set of different decisions, one must have the ability to distinctively categorize each choice.
To distinguish among different options within a set requires a vocabulary. And to choose among different actions requires, at the very least, different verbs. The existence of a grammar and vocabulary implies a language. Thus, the existence of conscious thought requires the ability to process language. QED.
This is just the beginning of course. Beyond this, we can introduce postulates about the existence of conscious thought, of different emotions, of different ways our conscious and unconscious interact, and about particular logical relationships, and derive even more truths.
We can even expand on our definitions, for example:
Observations can come via the external (sight, smell, etc.) or internal (memories, emotions, etc.) sources. We can then explore the mechanisms behind external or internal observation.
Since conscious analysis (aka logic) occurs via language, and since people think and speak in the same language, humans must have the same logic. And if we generalize language to Universal Grammar (see Noam Chomsky), then we can say that all humans use the same language, and therefore all humans use the same logic (see Mises on polylogism).
Decisions might feel consciously made but they may in fact be unconsciously made. (This is the hypothesis behind “System 1″ and “System 2″ thinking, popularized by Daniel Kahneman.) We may call these biases if we can overcome them. Or, if we can’t overcome them, we may argue that they are hard-wired limitations against free will.
An interesting situation arises with respect to babies and their state of consciousness:
I’m of the view that (1) babies are unconscious, and (2) they have an “ersatz” or unspoken vocabulary.
The evidence for (2) is that when children learn to speak language, they learn to categorize words with very few examples. They will know that a chair is a chair regardless if it’s made of wood or steel or plastic, regardless of its color or other aesthetic frills. In contrast, this is not how computers learn what a chair is, which is often by “training” on many examples labeled “chair”.
The simplest (although not only) implication is that the category of “chair” already pre-exists in the mind of a child. This goes beyond Universal Grammar, and perhaps suggests the existence of a Universal Vocabulary.
What this implies about Theorem 3 is that the language your consciousness mind communicates with your unconscious may not be the same as the language you speak with other people or yourself.
(Two side notes: first, the simplest way you can prove to yourself that you think to yourself in words, is that you can think of words that rhyme.
Second, the fact that the equivalent of the word “chair” may not exist in some languages does not negate any of the points above. The key is that any child born of any human parents is capable of learning to speak any language at the same rate as of other children.)
The evidence for (1) is that no one has memories of being an infant or embryo. I follow Mises (Human Action, chapter II section 2) in believing that memories are strictly a product of consciousness, and thus we should expect no memories from unconscious states. (This has interesting implications about remembering dreams.)
But following from my above point, since babies grow into their ability to speak, it may very well be that babies also grow into their consciousness. This is once again analogous to babies developing the muscle mass required to crawl.
This raises another interesting question: can we learn anything about the evolution of language, particularly its origin, from how babies develop?
The obvious inference is that consciousness developed gradually in humans as a species, like it develops gradually in a baby. Here’s one way of interpreting the historical record with this view:
Humans were first entirely unconscious animals; then circa 2 million years ago we slowly started developing consciousness which allowed us to use and develop primitive stone tools and fire, then circa 200,000 years ago homo sapiens appear, and about 50,000 years later we gained “full” consciousness that allowed us to use art and symbols.
However, there is an alternative way of interpreting the same series of facts: that consciousness developed all at once circa 50,000 years ago. Tool use is not indicative of consciousness; after all, robots can use tools, too. Insects also use tools, and they are often the prototypical example of automatons in nature.
Tool use among humans may have been entirely a function of a larger brain capable of more unconscious calculations, and having the body capable of holding and manipulating other objects.
This explains why, prior to advent of language, tool use developed very gradually — like other Darwinian changes.
Language use, however, arrived on the scene like a meteor. The historical record goes from having nothing that could be called art, to suddenly a veritable explosion of paintings, carvings, and sculptures. While we wallowed in the Stone Age for millions of years, within the last 50,000 we developed trade, money, math, writing, and civilization. Technology jumped from the Iron Age to Bronze Age to the Space Age.
Language use does not depend on our ability to speak, of course. We can still communicate with gestures, symbols, or even smoke signals. While it’s impossible to have a historical record of the evolution of gestures and smoke signals, the fact that visual symbols appeared all at once indicates to me that the ability to understand language appeared all at once, too.
I will have more to say about the evolution and origin of language in a future post. I believe even here we can use pure logic to shed light on how the process may have occurred.