Black Swans

The so-called “Black Swan” problem has always been one of my favorites. For me it represents one of those so-called philosophical problems invented by various sophists, such as Protagoras or Zeno. But the black swan problem is not really a sophistic argument at all. It is simply a huge mistake. The wonder of it is that so many otherwise brilliant minds could have been taken in by it.

The mistake rests on three other mistakes: bad ontology, the philosophical mistake called induction, and the misunderstood relationship of epistemology to science.

The reason the black-swan fallacy is important is because it is used today in many fields to justify all kinds of wrong views just as two arguments of Protagoras, “everything is relative,” and, “there are two sides to everything,” continue to be used to put over wrong ideas. Nassim Nicholas Taleb, for example, is an economist who uses his own variation of the black swan fallacy as the basis for his own theory of economics.

Black Swans

Why are black swans a problem to science? Of course they are not, but the so-called problem is based on the mistaken idea of induction. It means no number of observations can ever result in a universal generalization because, even if all the swans that are ever observed are white, the generalization that “all swans are white” can be falsified (proved untrue) by the observance of a single black swan. The black swan is used as the example because it was believed in Europe that all swans were white because only white swans had ever been observed, until black swans were discovered in Australia, by Willem de Vlamingh in 1696.

No Such Thing As Induction

Observation is the beginning of all scientific discovery. The purpose of science is to discover the nature of what is observed in the physical world, the things (existents) and events that are observed and the relationships between them.

How the mistaken idea that scientific discoveries are made by some method of repeated observation called induction involves the entire history of philosophy and science, beginning with the arch antiphilosopher, Hume. There is only one method of correct reason, deduction, which is formalized as deductive logic.

The true method of science that has been twisted into the wrong view of induction, is the epistemological method of observation and identification. [Epistemology is the philosophical study of the nature of knowledge.]

All knowledge is by means of concepts (words)¹. Concepts have only one purpose or function, a concept identifies an existent (particular) or class/category of existents (universals), including all entities and other concepts, their attributes, events (the action of existents) and their relationships.

What Is An Existent

Ontologically, only entities exist primarily. Events, attributes, and relationships exist but only secondarily—events are the actions of entities, attributes are attributes of entities, and relationships are between entities or their actions and attributes. There are no “wild” events, attributes, or relationships independent of the entities of which they are the behavior, qualities, or relationships. [Ontology is the philosophical study of nature of existence.]

It is an existent’s attributes or qualities that determine what it is. An existent’s attributes do not make it what it is, its attributes are what it is. To identify an existent’s attributes is to identify the existent. A concept therefore identifies an existent by means of it attributes or qualities (which includes its behavior and relationships), and means that entity with all its attributes and qualities (because that is what it is). The concept’s definition does not need to specify all an existent’s attributes, it only needs to specify as many as are necessary to distinguish the existent from all other known existents.

When a scientist, or anyone else, observes a large white aquatic bird with a graceful s-shaped neck and golden bill and chooses to use the word “swan” as the symbol for the identification of (concept for) that bird it is an example of what science does—observation and identification. Notice there was no need of multiple observations of the bird for that identification. Notice that the identification cannot be wrong because the attributes the bird has are its identification.

If some individual observes another bird with the same attributes it will also be correctly identified as a “swan.” In fact, every bird ever observed with the same attributes will be correctly identified as a swan, because it is their attributes that are their identification as the kind of existent they are.

Universals Established By Conceptualization, Not Induction

Universal generalizations are not established by multiple observations at all, they are established by means of forming concepts to identify existents by means of their attributes. To say all swans are white is correct because that is part of their identification. What makes the concept universal is that it subsumes all birds with the same attributes, that is, the same identification.

The discovery of a bird, similar in all respects to a swan with one different attribute (like color) does not cancel or negate or falsify the identification of swans as white. The new bird is just that, a new kind of existent, requiring a new identification and a new concept.

Because it is so similar to a swan in all other respects, the question of what word ought be used to identify it is not a scientific question but an epistemological one. It might be considered a subset of the concept for swans, or it might be considered a separate kind of existent. The right name for that epistemological field is taxonomy, not science!

When the nature of scientific identification is made clear, the obvious absurdity of the so-called “black swan” argument becomes obvious. The term “swan” was never an argument about what could or could not exist or be true, it was the identification of what was known to truly exist. The discovery of what came to be called “black swans” did not change that true identification. If it did, it would not be necessary to use the phrase, “black swan,” to distinguish the newly discovered birds from the already identified swans.

The true nature of scientific observation and identification also clarifies why induction has never been any part of true scientific discovery, and anything that is put over as science on the basis of any such inductive method is not science.

The Example Of Chemistry

John Dalton’s research that would eventually lead to the multiple proportions hypothesis in chemistry (now established as a principle or theory) involved the identification of many different chemical compounds, among them were the various oxides of nitrogen.

The Oxides of Nitrogen

The table illustrates the ratio of nitrogen to oxygen of each compound that resulted from Dalton’s research. What I want to emphasize from the table is the first step in the research which was the identification of the different compounds themselves.

Each of the different compounds of nitrogen and oxygen were discovered because each had their own unique attributes. As each was discovered it was identified in terms of whatever was known about them at the time of their discovery (e.g. color, odor, state {solid, liquid, gas}, any known reactions with other known elements or compounds, etc.).

Once identified, further experiments revealed additional attributes. That these were all compounds of the same two elements was not discovered until they could be, “broken down,” and their constituent elements (nitrogen and oxygen) identified.

At no time was induction used as any part of the discovery of the nature of these compounds. The discovery that nitrous oxide and nitric oxide were both compounds of the same two elements was not established on the basis of multiple observation. When either compound was broken down, only oxygen and nitrogen were found. In fact, it only needed to be observed once. Certainly, multiple experiments were done to ensure the observations were correct and that there hadn’t been any error (that’s called rigor, not induction). If any experiment had produced another element in addition to nitrogen and oxygen, the correct observation and conclusion would have been that another compound had been discovered, because it had attribute the other compounds did not.

Wrong Views Proliferate

The black-swan fallacy is based on bad philosophy and bad science. The problem with views that are both wrong and widely accepted, is that in whatever field they originate they end up infecting many others. One of the best examples is Nassim Nicholas Taleb’s “Black Swan Theory,” of economics, and almost everything else. He said so himself:

“A small number of Black Swans explains almost everything in our world, from the success of ideas and religions, to the dynamics of historical events, to elements of our own personal lives.”

In any field, something that proposes to explain everything, usually explains nothing. This is certainly true of Taleb’s Black Swan Theory.

A black swan or black swan event, according to Taleb is a rare or unpredictable event that is nevertheless very important or has a, “high impact.” Taleb is an economist and as most economists do, tends to view everything in statistical terms. When Taleb describes the three characteristics of black swan events, it is in terms of statistical probability. The three characteristics of black swan events he says are:

1. The event is a surprise because it was unpredictable.

2. The event has a major effect or massive impact.

3. After the event, it is rationalized that it could have been expected or predicted.

Every discovery is a future event, something that becomes known, or known about, that was not previously known. Every discovery is a surprise. It is a surprise precisely because it was not expected or predicted.

The future is unpredictable. It’s what makes it the future. There are real factors which make that true, especially if what one attempts to predict includes human behavior. The entire physical world is unpredictable because it is too big for all possible determinants and relationships to be known, too complex for every possibly variable to be identified, and all mathematical prediction is limited by incommensurables (irrationals and transcendentals) and by all events that have fractal determinants.

When any human component is included in an attempt to make predictions, reliable predictions are impossible. No human behavior can be reliably predicted, because every human being is a unique volitional being and therefore behaves in no pre-determined way.

So every future event is a surprise to some degree, but so is how much “effect” or “impact” any event will have. 1 and 2 should be combined to say, a black swan event is an event with a big unexpected effect. That describes just about every discovery and significant event in the history of the world, from those of the explorers to those of the sciences and every technological development. It was nice that Taleb got around to noticing their significance.

The third characteristic is really irrelevant. No amount of rationalization can change the unpredictability of the future. I’ll make one other observation about it, however. Since statistics supposedly makes predictions on the basis of probability, Taleb’s theory essentially knocks statistical prediction on its head. Any event that actually occurs had a one hundred percent possibility of occurring, but no statistical method could ever have predicted it.

—(03/10/16)