Content-type: text/html ~ Stephen's Web ~ What I've Learned From Philosophy

Stephen Downes

Knowledge, Learning, Community

Half an Hour, Jul 30, 2015

I posted an item in OLDaily today from Forbes touting the benefits of formerly 'useless' liberal arts degrees. In this item Slack CEO Stewart Butterfield is quoted:

"Studying philosophy taught me two things," says Butterfield, sitting in his office in San Francisco's South of Market district, a neighborhood almost entirely dedicated to the cult of coding. "I learned how to write really clearly. I learned how to follow an argument all the way down, which is invaluable in running meetings. And when I studied the history of science, I learned about the ways that everyone believes something is true–like the old notion of some kind of ether in the air propagating gravitational forces–until they realized that it wasn't true."

It's worth mentioning because the Department of Philosophy at the University of Calgary has a notice posted on the wall to the effect that a philosophy degree was no guarantee of a job and that graduates should not study expecting employment in the field. This wasn't parody or in any way humorous - it was an official memo from the chair and posted in all seriousness. I can still see it in my mind, not a big poster but an 8.5 x 11 memo with typed text.

I have often commented that my work in philosophy left we particularly well-suited to employment in the new economy. It's not merely that sorting out corporate information might be simple after spending years teasing out the nuances in Wittgenstein, as the article suggests, though it's partially that. It's about what it is to know and to learn at a deeper level, which can then be applied to new disciplines whatever they may be.

But what, precisely, did I learn from those years of study? That's a hard question to answer. But it's worth a bit of a sketch here.


Butterfield said he learned to write clearly. But what does that mean? Fun with Dick and Jane is written clearly but we want to express thoughts more complex than "see Spot run." Writing clearly means writing with precision, and precision is what philosophy teaches.

For example, it is commonly said that a sentence has a subject and a verb. This proves to be important in clear writing. In clear writing the subject of the sentence is unambiguous. The reader knows exactly what you are talking about. Through the rest of my days I have always been attentive to the identification of the subject. You would be surprised how many people are not.

There are specific ways of naming the subject. One way is to point, in words (that is, to name your subject ostensively). "This is a sentence. That was an argument worth hearing." Wittgenstein did that a lot. Another is to use a definite description. "The present King of France,"  for example, was the subject of much discussion between Russell and Strawson. Another way is to use names, which may in turn be subject to definitions, for example, "dogs", "millennials" or "Barack Obama".

How many ways are there to be imprecise about the subject? There is always our favourite case, the amphiboly:  "One morning I shot an elephant in my pajamas. How he got into my pajamas I'll never know." Another is to refer to something without a definite or indefinite article, for example, saying "Matter of importance is clarity," instead of "A matter of importance..." or "The matter of importance." Or there is the use of vague terms: "freedom is what defines our approach to software." And on and on the list goes.

Precision is what lies at the root of grammar. In my opinion, the rules of grammar (for the most part) exist in order to ensure precision. A lot of times it is the little things that cause confusion. A single comma can change the entire meaning of a sentence. As when Johnny said, "It's supper time. We're ready to eat, Uncle Charlie."


What you learn in philosophy is that sentences - and thoughts generally - are not unstructured streams of consciousness. This is especially clear in languages like French, where you have to plan your sentences ahead of time, in order to ensure the gender of your words are in accord. In all languages, structure indicates not only the subject and verb, as mentioned above, but also logical form leading to such things as inference and explanation.

Why does this matter? Well, as I've written elsewhere, understanding this structure is key to writing useful and meaningful essays. It is also key to being able to analyze and understand what other people have written. When you read an editorial containing a whole list of sentences, how to do determine what opinion they are trying to express? It is the structure of the article that tells you this.

Structure is logic, and logic is structure. You can see this by looking at the different kinds of logic; they reveal to you the different kinds of structures you can employ in your reasoning:

  • propositional - connecting and relating the truth of basic sentences using 'and', 'or', 'if-then' and 'not'.
  • quantificational - specifying how many of something we're talking about, and inferring about properties of groups of things
  • causal - understanding the conditions under which one thing is said to cause another
  • modal - talking about whether things are 'necessary' or merely 'possible'
  • statistical - understanding probability, that is, how likely something is to happen, or to be true
  • deontic - thinking about the nature of obligation and permission
  • doxastic - the logic of beliefs
  • mathematical - axioms, calculus and set theory 
  • computational - Turing machines and computational processes

Not only did I learn that all these forms of logic exist (who knew?) I also actually learned them, which means I can make really complex inferences, but more importantly, know some pretty basic things. For example, if 'P' is necessarily true, is 'P' true? (Yes) Or for example, if 'P is always Q' is true, does it follow that 'P' is true, or that 'P' exists? (No).

Syntax and Semantics

Syntax is the structure of something - its logic - while semantics refers to its meaning, truth or value. Syntax is the fact that ten dimes make up a dollar; semantics is the fact that it takes ten dollars to attend a movie.

The very fact that syntax and semantics are distinct is important in itself, for several reasons.

The first is that syntax is arbitrary. We can make up any sort of syntax we want. This is not so easy to see in everyday arithmetic and propositional calculus, where the rules are deeply entrenched. But in modal logic, however, we have various 'systems' such as T, K, S4 and S5. Which one of these is 'true'? Well, they all are. Or none of them is. Or, it doesn't even make sense to ask the question.In mathematics, similarly, there are different axiom systems. Which is 'true', Peano arithmetic? Mill's Axioms? Or does it even matter?

In fact, a syntax, thought in and of itself, can be whatever we want it to be. Usually we set out some basic requirements - the system should not allow contradictions, for example. But there's no requirement that we do this, and if we develop a system that does not have truth as its basis (language, say) then the principle of non-contradiction doesn't even make sense! Take a look at my categorical converter - do the lines have to be drawn that way? Well, no. Or imagine a logic that is falsity-preserving, rather than truth preserving: they look like mirror images, but in falsity-preserving logic, nothing follows from a contradiction, and everything follows from a tautology.

If pressed, we would say that we need to choose one system of logic over another because one of them works in the real world, but the other doesn't. But the relation between logic and the world is far from clear. We 'prove' a system of logic with a semantical argument, but the relation between a semantics and a logic is itself the subject of discussion; these different relations are called 'interpretations'.

What does it mean, for example, to say that "the probability of 'P' is n"? There are three major types of interpretations of this statement:

  • the logical interpretation, from Rudolf Carnap - for every possible state of affairs in which P could be true or false, in n of them, P is true.
  • the frequency interpretation, from Hans Reichenbach - in all cases in the past where P could be true or false, in n of them, P is true
  • the subjectivist interpretation, from Frank Ramsay - of you were to make a bet on the likelihood that P is true, you would require odds of n

So if we are to validate the laws of probability - Bayes Theorem, for example - against an empirical model, which of these is the correct model to choose?

For that matter, what makes a statement P 'true' at all? Alfred Tarski said "the sentence 'snow is white' is true if and only if snow is white." Well, that sounds good. But the sentence "brakeless trains are dangerous" can be true even of there are no brakeless trains. So it seems there are two basic principles of truth - a correspondence principle, which requires reference to a physical world of some sort, or a coherence theory, which requires consistence with a model.

This is how murky these questions can get when we're talking about something as basic as truth. In the 20th century, however, philosophers focused on other aspects of semantics, such as meaning and value. Here, the discussion became even more murky.

When someone comes to me and says that some thing or another is 'true', you can see I have a lot to think about regarding what this assertion could possibly mean. When somebody says to me that "We can all agree that such and such," I begin to distrust this person, first because the statement is probably false, and second because it's not at all clear to me that 'agreement' is even relevant to the sort of truth, value or meaning that we are discussing.

These are really important lessons, and they apply everywhere.

What are 'Things'?

Philosophy taught me that anything can be a 'thing' - it just depends on how you look at it. And that there are different types of things, and different types of types of things.

Our teachers in school spent a lot of time telling us about the basic types of things - animals, minerals and vegetables - and the different types of each thing that fall neatly into categories beneath them as kingdoms, phyla, species and genera.

In university I learned that the way we define a thing in this system is to identify the category a thing belongs in, and what distinguishes it from other members of that category. "A cat is a mammal that purrs." "A hammer is a tool used to drive nails." That sort of thing. "An x is such that all x are y and only x are z." Necessary and sufficient conditions. Essences.

Then in philosophy I learned that all of this is arbitrary. The beautiful system was upended, most notably, by Wittgenstein. "What is a game?" he asked. Is there any statement that is true about all games? No. Is there any statement that is true about only games? No. The idea of a 'game' os that it is a bunch of things that are kind of the same, like family resemblances, so you can see that they are sort of alike, but there is nothing unique that defines them.

Language itself is like this. We don't have 'rules' properly so-called, we have "language games". What does a word mean? Well, it depends on how we use it. The meanings of words, the rules of language, the nature of what is true and what isn't - these all shift over time, like the bed of a river.

Even more importantly., what a thing is depends not on the thing itself, but on how it is observed. Because whether one thing 'resembles' something else really depends on your point of view. We can in one sense say that checkers resembles chess, while in another sense say that checkers resembles mathematics.

There are many ways to define things: we can point to them, we can say what they contain, we can say what properties they have, we can talk about what they do, what they were designed to do, what they actually do, what they might do, we can say what they're for, we can talk about where they're from or who (or what) created them, and on and on.

Viewed this way, anything can be a 'thing', and any group of things can be a thing. George Lakoff talks about the culture that divides the world into two types of things: one class consisting of  "women, fire and dangerous things," and another class consisting of  everything else.

So much of what we do today involves either working with certain types of things, or understanding that we are defining new types of things. What are 'students'? What is a 'learning object'? How do we define an 'ontology'? Philosophy taught me about the limitations of relational databases long before there were relational databases.

Theories and Models

Quine's Two Dogmas of Empiricism taught me (and everyone else) two important things:

  • There's no such thing as the analytic-synthetic distinction
  • Reductionism is false

Above I discussed the distinction between syntax and semantics. The collapse of the analytic-synthetic distinction means that no statement is wither purely syntactical or purely semantical.

What does this mean? An analytic statement is supposed to be true simply by virtue of the meanings of its terms. We say "1 + 1 = 2" is true, not because of some fact about the world, but because of the meaning of the terms '1' and '2' and '+' and '='. But if we put it this way, no statement is purely analytic. "It is obvious that truth in general depends on both language and extra-linguistic fact."

This leads us to the second dogma: reductionism. This is the idea that all true statements can be reduced to 'observation language' or some other basis in pure facts (this could be any set of facts: facts about the world, facts about pure thought, facts about the Bible). But in fact, there is no set of 'observation statements'. Every 'fact' carries with it some element of the theory it is purporting to prove. For, without the theory, there is no way to say whether even a simple sentence like "the sky is blue" is true or false.

This taught me, critically, that what a person sees depends on what that person believes. It means we have to rethink how we approach research and discovery, but also that we have to rethink how we communicate with people, how we appeal to reason and evidence, and even how we regard the world and our place in it ourselves. And it's why education - and how we think of education - is so important.

For example, I say "to teach is to model and demonstrate". These are not idly chosen concepts. What we model impacts how they see the world. Consider four world views (all of which correspond loosely with different generations in and around my lifetime):

  • We're at war. Our heroes are war heroes. When we work, we're at the front line. The challenges we face are battles. The determination of a Churchill or a Patton inspire us.
  • We are explorers. We use science and technology to discover new things. When we work, we are solving problems. The challenges we face are mysteries, the unknown. The courage of John Glenn and James T. Kirk inspire us.
  • We are players. Our heroes are athletes who bring out the best in themselves. We leave it on the playing field, but experience camaraderie outside the arena. The strength of Gordie Howe or Hank Aaron inspire us.
  • We are entrepreneurs. We take ideas and make change in the world, bending vast empires of money and people to our will. We are driven by results, and expect a return on our investment. Our heroes are people like Bill Gates and Steve Jobs.

And there are many more, in different generations of different societies around the world. Each of these does not represent just a different world view or a different paradigm. It represents a different way of life. Without philosophy, it's impossible even to understand that there are other ways of life, much less to understand what they could be like.

What are 'evidence' and 'proof' to people in each of these different worlds. I inhabit a workspace where the only measure of whether something has value is whether someone will pay for it - part of that entrepreneurial mindset. I don't agree with that mindset, but I'm also aware that my own mindset, the explorer mindset, isn't inherently superior.

People are always saying to me that "this counts as a theory, but that doesn't," or that "this counts as research, but that doesn't." I recognize such statements as arbitrary, and representing a set ofd parameters that the speaker has employed to define what will count as 'normal' (or 'standard', or 'appropriate') in their lives and work. I know I won't change their minds on this, probably, because no evidence exists that does not reinforce their world view. That's the nature of world views.

Thought is Associative

Not everybody who studies philosophy will learn this (see the preceding paragraph) but I did, and it was of fundamental importance to me.

There are different ways to make the same point. Other people, for example, will say that they learned that not everyone is rational, or that people don't make rational decisions. Others will say that people think in music and pictures and whatever. These are both true. But for me, it comes down to the idea that thought is associative.

But what does it mean? It's hard to explain in words, but by way of a metaphor, I would say that the principles of knowledge, memory and understanding are basically the same as the principles that apply when you throw a rock into a pond. There is the impact, there is the cascade as waves rush out from the rock, there is the pushback as waves bounce off each other and off the shore, and there is the settling as the pond returns to its level.

Now the human brain is much more complex than a pond, but in both cases, the impact of something new affects the entire system, even though the cause touches only one small part of it. The rock touches some water, which pushes against other water, which pushes against a shoreline, and so on. The water organizes itself through a whole series of molecule-to-molecule interactions. There's no head molecule. There is no 'purpose' or 'order' defining what the waves must be - if tyhe stone had been bigger, the water colder, the shoreline shaped differently, it would have worked out in a completely different way.

We are on the verge of understanding how that process actually works in brains (we understand pretty well already how it works in ponds, to the point that we have an entire discipline built around fluid dynamics). What we don't have yet is a way of understanding the world consistent with this understanding of how thought works.

For example, I have said frequently, knowledge is recognition. Water doesn't really retain the impact of rocks, which is why ponds aren't intelligent. But other more complex and more stable entities will retain traces of the impact. One thing influences the next, and each thing preserves a trace of that influence, such that after a while characteristic patterns of input produce characteristic responses. This is recognition. And it is, to my mind, the basis for all human intelligence.

This way of thinking is in an important sense post-semantic. I don't see one thing as a 'sign' for another. I don't see mental models as 'representations' of some external reality. I see knowledge, cognition and communications as complex interplays of signalling and interaction, each with no inherent meaning, but any of which may be subsequently recognized by one or another entity.

Remember how Marx said "everything is political"?  Well, I think that "everything is a language" (or, alternatively, there's nothing special about language over and above other forms of communication). So when I create a 'scientific theory', which is my job, I create something that consists of language, code, actions, photographs, and a host of other artifacts, all of which are reflections of my interactions with the world, not intended to 'represent' some deeper truth or underlying reality, but rather, intended to offer a set of phenomena that may be usefully employed by others (depending on what they recognize it as being useful for).

Born Free

In any number of recent movies - the Hunger Games, for example, or Divergent - the plot revolves around the idea that society is structured in such a way that we all have our assigned places where we work and live. Sometimes, as in Harry Potter, this is depicted as a good thing. But more often the established order is the subject of resistance.

The concept originates in Plato, who in the Republic argued that society should be run by philosophers, and that the position of each person would be determined by their inner nature. "One man will acquire a thing easily, another with difficulty; a little learning will lead the one to discover a great deal; whereas the other, after much study and application, no sooner learns than he forgets; or again, did you mean, that the one has a body which is a good servant to his mind, while the body of the other is a hindrance to him."

It is true that there are innate variations among humans. But the far greater differences between people are the result of their upbringing, culture and education.

In philosophy I encountered the idea that there is an inborn 'human nature' on a regular basis, from the above-mentioned assertions from Plato to Descartes's ideas about the stamp of God implanted in the human brain to Chomsky's postulation of an innate deep grammar. People argue that there are common things (love of justice, fear of death) that unite us all, and essential properties (mental capacity, physical strength, mathematical abilities) that divide us.

But none of this is true. What we have in common operates at a far lower level than people suppose. It operates at a genetic level, a cellular level, which defines only the most basic principles of human composition. Our heritage determines that we will have leg muscles, but not how string those muscles will be. It determines that we have interlinked neural cells, but not how they will be wired together. It determines that we will have a voice, but not what we will say.

Time and time again I have encountered evidence of this. When we look at physical properties, for example, and even the oft-touted difference between men and women, we see how large a role nutrition plays (women are tall and strong in nations where they are well-fed and nourished - think about that). The physical differences between individual members of any race, class or gender you can to name are far greater than any between the races, classes or any other identifiable group.

The same is true of mental properties. Time and time again, the most reliable predictor of educational outcome is socio-economic status. This is not because (as some suggest) the best and brightest become rich (surely we have countervailing evidence of that) but because of the advantages they receive in early life, everything from a rich intellectual environment, proper nutrition and stimulation, and social expectations supporting learning and achievement.

How much of philosophy is devoted to determining whether there are natural - or essential - properties of things, and most especially humans? Arguably, most of it. The argument that something 'must be X' on the basis that 'X has property or capacity Y' runs through the entire history of philosophy, from Thales to Aquinas to Kant to Fodor. And none of these speculations has ever stood the test of time.There are no innate properties of significance. We are born free.


The word 'value' is a bit loaded as in our entrepreneurial age it has become virtually synonymous with some means of quantification in terms of worth, utility or commodity. There are older senses in which the term 'value' meant something like those properties synonymous with virtue, but those senses of the word are almost inaccessible to us now; we would have had to have been born in a different time and a different place to understand it.

I think philosophy has taught me to think of value a bit more deeply than that, and to at least be able to articulate alternatives that can count as 'value'. These alternatives form the basis of the various systems of morality and justice that have prevailed over the years.

I once wrote to the Globe and Mail in a no-doubt long-lost online forum that the underlying value that defines Canada is this: in diversity, harmony.

You need both parts. Harmony is the underlying value (the earth, as the Taoists might say), the receiver of all things, the pond after it has become stable, the mind after it has become calm, uncertainty and turmoil resolved. But rocks and sand crabs and fungus also exhibit large degrees of harmony; we want something more. This is provided by diversity, the possibilities of experience, the creation of the need to adapt, to understand, to grow and to learn.

But hey - it's just a value system. It's not like others haven't tried before me. And this knowledge keeps me humble.

One type of value system revolves around survival. It's an animal value system, an artifact of our lizard-brain, perhaps, brought through centuries of socialization to mean also the survival of the offspring, survival of the tribe, or survival of the species. We see it reflected to day in such philosophies as social Darwinism, survivalism, and various types of rule-based tribalism.

Another type of value system revolves around ideals. We have the Platonic forms, the perfect Christ, Man and Superman- the idea is that the closer we can come to perfection, the greater the value we have realized.

Another type of value system is based on duty and obligation. Perhaps best represented by Kant, it is informed by the idea that each person is an "end in themselves", not a means to an end (today we would say "each person is inherently valuable") and that we ought to act in the manner such that every person could also consistently act in the same manner. Your mother invokes Kant's categorical imperative when she says, "What is everyone else did that?"

Still another is based around the idea of happiness, and of freedom in a manner that enables a person to maximize their own happiness. People like Jeremy Bentham and John Stuart Mill are most closely associated with this philosophy, and Mill famously proposes that the goal of society ought to be to allow each person to pursue their own good in their own way. I have a lot of sympathy with that ideal.

Maybe they all amount to the same thing. There's no shortage of ecumenical authors who like to suggests that, at heart, we all have the same system of values. But if this were true then we would have no satisfactory explanation for a Jeffrey Dahmer or a Clifford Olsen. So even while it feels to me that hose perfect moments of harmony are a combination of happiness, obligation and ideals, I think that other people see these values very differently.

This is important to understand. People like to say things like "the truth lies somewhere in the middle" or "the good is what we can all agree on". But there really is no such thing (or if there is, we have utterly no means of finding it just yet). 


I was never really a fan of moral philosophy, because of the force of the observations just presented, and even less of political philosophy, which to my way of thinking was offered for the most part by the powerful to rationalize their exercise of power.

Of course, I have probably been jaded by the fact of being born and raised in an environment where the peak of political philosophy varied between people justifying why we would have enough military might to destroy the entire planet and people giving reason why we should or should not use it. Political philosophy in my age is and continues to be about the deployment of political power.

Probably the predominate idea in political philosophy is some sort of version of social contract theory. This is the idea (and we see it reflected in school charters and corporate vision statements) that we are united as a society under a set of principles that we have agreed to in order to live together, prosper together or learn together.

The motivation for such a social contract is generally that the alternative is unbearable. Without, for example, the benign power of an absolute sovereign, wrote Thomas Hobbes, our loves would be "solitary, poor, nasty, brutish and short" (or course, given some of the sovereigns he was defending that might be preferable).

The idea that we have actually signed such a contract is, of course, absurd. So the nature and standards of conduct in the contract are often implied - Rousseau, for example, appeals to the state of nature in which the noble savage found himself, as compared to contemporary society - "man is born free, but everywhere he is in chains." John Locke, envisioning an endless commons, imagines that the rights of property are established when someone "affixes his labour" to that which may be found in nature (an argument that justified the conquest of North and South America). John Rawls imagines that we could imagine what we would negotiate with each other under a "veil of ignorance" in which no one knew whether they would be a pauper or a king; this would result in a system of "justice as fairness".

And of course there are communitarian theories of  political philosophy based around the common ownership of "the means of production" which would ensure that everyone gets "to each, according to his needs, from each according to his means." Other communitarian theories of justice assert the collective rights of women, minorities, language groups, religions, and others.

Interestingly, I don't think that anyone who is actually in politics subscribes to any of these philosophies per se. Actual observation  (if there is such a thing) suggests that most of our social and economic leaders are engaged in one or another version of Machiavellian political theory, loosely stated as "might is right".

For my own part, I don't know whether "man is born free," but I do observe that "everywhere he is in chains," and just as I feel the limitations of my own self-actualization I feel that the other people of the world who have even less advantage than I do must feel more or less the same thing, perhaps more deeply. I do not see us as merely "workers" or even as members of this or that community; from Kant I draw the idea that each person is equally important and equally special, and that our society and our individual lives are most enhanced by realizing that.

But I have no illusions, I don't believe in utopia, and I don't believe we can engineer (as so many political philosophies suggest) a better society, a better company or a better school. In the end, the political philosophy we employ - the nature of our culture, our social believes, our nation - is the result of a billion individual, decisions made every day, and each of these decisions is based on the many factors I've outlined above.

Good government, in other words, depends as much on things like precision of language, structure of reasoning, appropriate semantics, and all the rest, and even then, there's no guarantee that the government we get will be in any meaningful sense good - the best we can hope for, maybe, is government that is just, and leave the rest to the people.

In Sum

In sum, philosophy has taught me the basics of what I need to conduct myself in virtually any enterprise or occupation (save perhaps things like Major League Baseball).

I've learned through philosophy that nobody is special, and everyone is special. That nothing is real, and everything is real. That there infinite ways we can describe and divide up the entities in the world, that in practice we fall into habits of seeing and reasoning about the world based on our experiences and the influence of those around us (and today, that influence includes language and media).

I think that the reason we are alive is because it's possible, and the reason we die is to continue to allow it to be possible, by allowing our form of existence to grow and develop and adapt and flourish.

I'm still trying to embrace diversity, and I'm still seeking harmony.

Stephen Downes Stephen Downes, Casselman, Canada

Copyright 2024
Last Updated: Jul 22, 2024 10:08 a.m.

Canadian Flag Creative Commons License.