Essay Reality Truth

On By In 1

Truth, like knowledge, is surprisingly difficult to define. We seem to rely on it almost every moment of every day and it's very "close" to us. Yet it's difficult to define because as soon as you think you have it pinned down, some case or counterexample immediately shows deficiencies. Ironically, every definition of truth that philosophers have developed falls prey to the question, "Is it true?"

Simply, we can define truth as: a statement about the way the world actually is. We'll look at various theories below that philosophers have considered but that's an adequate rough-and-ready definition to get us started. Coming up with a definition of truth falls under the discipline of epistemology or the study of knowledge though some philosophers categorize it as a study in metaphysics--the study of what is real.

In this essay, we'll look at some reasons why defining truth can be challenging. Truth seems like something we naturally comprehend and while intuition can help us a great deal in understanding what it is, surface definitions present us with unique problems and I’ll illustrate why. I'll then lay out some terms and concepts that will help us get a better handle on understanding what truth is. Next, we'll look at three main views of truth. The coherence theory describes truth in terms of interconnected belief. A belief is true if it is consistent with other beliefs we have. The correspondence theory describes truth in terms of a relation concepts or propositions have to the actual world. Finally postmodernism lays out a view of truth in terms of individual perspectives and community agreement. While this essay does not focus on practical issues like why a view of truth is important, I'll say a few words about that idea at the end and provide more resources for further reading.

Elusive Truth

I stated above that defining truth can be challenging. Let’s briefly look at why this is so by way of a seemingly simple example. Suppose you examine an apple and determine that it’s red, sweet, smooth and crunchy. You might claim this is what the apple is. Put another way, you've made truth claims about the apple and seemingly made statements about real properties of the apple. But immediate problems arise. Let's suppose your friend is color blind (this is unknown to you or her) and when she looks at the apple, she says that the apple is a dull greenish color. She also makes a truth claim about the color of the apple but it's different than your truth claim. What color is the apple?

Well, you might respond, that's an easy problem to solve. It's actually red because we've stipulated that your friend has an anomaly in her truth-gathering equipment (vision) and even though we may not know she has it, the fact that she does means her view of reality is incorrect. But now let’s suppose everyone is color blind and we all see "red" apples as green? We can make this objection even stronger by asking how we know that we all aren't in fact color blind in a way we don't understand and apples really aren't red after all. No one has access to the “real” color of the apple. Again, the response might be that that this is a knowledge problem, not a truth problem. The apple really is red but we all believe it’s green. But notice that the truth of the apple’s color has little role to play in what we believe. No one knows what the truth is and so it plays no role in our epistemology.

The challenge is that our view of truth is very closely tied to our perspective on what is true. This means that in the end, we may be able to come up with a reasonable definition of truth, but if we decide that no one can get to what is true (that is, know truth), what good is the definition? Even more problematic is that our perspective will even influence our ability to come up with a definition! These are no small concerns and we'll explore some responses below.

Some Preliminaries

Before we get to definitions of truth, we need to define some terms used in those definitions which will make things a little easier to digest. Epistemologists (people who study truth, belief and knowledge) use the following concepts as the framework for their study of truth.

Propositions. A common technical definition of a proposition (credited to Peter van Inwagen) is "a non-linguistic bearer of truth value." A proposition is a representation of the world or a way the world could possibly be and propositions are either true or false. Propositions are different than sentences. Sentences are symbolic, linguistic representations of propositions. Okay, that's all very technical. What does it mean?

Let's take the sentence, "The moon has craters." This is an English sentence that supposedly states some fact about the world or reality (and specifically about the moon). Because it’s in English, we say it's "linguistic" or language-based. If we're going to get philosophical about it, we could describe its properties as having four words and 17 letters, it's in the English language written in 11 point font and it's black. I could write the same sentence like this:

The moon has craters.

This sentence has different properties from the first one above. This one still has the same number of words and letters and it's in English. But it is in 18 point font and is written in blue. Now let's take this sentence, "La luna tiene cráteres." This sentence has four words but 19 letters. It's written in 11 point font and is black but it's Spanish. What do all three sentences have in common? Well, they all express the same idea or meaning and we could say the same "truth." We could express the same idea in Swahili, semaphore, Morse code, or any other symbolic system that conveys meaning.

Notice that the symbols themselves are neither true nor false. The meaning the sentences represent is either true or false. Sentences are symbolic representations of something else—propositions. The common property true of all sentences that express the same truth is what philosophers call the propositional content of the sentences or "the proposition." Now we can better understand the idea behind "non-linguistic bearer of truth value." Propositions are non-linguistic because they aren't written or spoken in a language. They bear truth because they are the things that are true or false. This is what allows them to be expressed or "exemplified" in a variety of different symbolic systems like language-based sentences. When it comes to understanding truth, many philosophers believe propositions are at the center.

Belief. Beliefs are things (at least) people have. They don't exist outside the mind. Some philosophers say beliefs are "dispositional." That is, they incline a person to behave in a way as if the thing they believe is true. So a belief, simply, is a proposition that a person accepts as representing the way the world actually is. Beliefs can be about false propositions and thus be "wrong" because the person accepts them as true. This is a critical distinction. While a proposition has to be true or false, beliefs can be about true or false propositions even though a person always accepts them as being true.

Some philosophers attempt to define truth "mind-independently." That means, they want to come up with a definition that doesn't depend on whether humans can actually believe or know what is true. Truth is viewed as independent of our minds and they seek a definition of it that captures this. Other philosophers have developed theories that keep people at the center. That is, truth and belief are considered together and are inseparable. I will try to make the relevance of the "epistemic" vs. "independent" views of truth relevant below.

Knowledge. Knowledge is belief in a true proposition that a person is justified in holding as true. The conditions under which a person is justified is complicated and there are many theories about when the conditions are met. Theories of knowledge attempt to describe when a person is in a "right" cognitive relationship with true propositions. I describe some theories of knowledge and some of the challenges in understanding when a person knows in an article for Philosophy News called "What is Knowledge?"

Common Definitions

The Coherence View of Truth

The main idea behind this view is that a belief is true if it "coheres" or is consistent with other things a person believes. For example, a fact a person believes, say "grass is green" is true if that belief is consistent with other things the person believes like the definition of green and whether grass exists and the like. It also depends on the interpretation of the main terms in those other beliefs. Suppose you’ve always lived in a region covered with snow and never saw grass or formed beliefs about this strange plant life. The claim "grass is green" would not cohere with other beliefs because you have no beliefs that include the concept "grass." The claim, "grass is green" would be nonsense because it contains a nonsensical term "grass." That is, you never formed a belief about grass so there’s nothing for this new belief to cohere with.

As you can see from the above description, coherence theories typically are described in terms of beliefs. This puts coherence theories in the "epistemic" view of truth camp noted above. This is because, coherence theorists claim, we can only ground a given belief on other things we believe. We cannot "stand outside" our own belief system to compare our beliefs with the actual world. If I believe Booth shot Lincoln, I can only determine if that belief is truth based on other things I believe like "Wikipedia provides accurate information" or "My professor knows history and communicates it well" or "Uncle John sure was a scoundrel".

These are other beliefs and serve as a basis for my original belief. Thus truth is essentially epistemic since any other model requires a type of access to the "real world" we simply can't have. As philosopher Donald Davidson describes the situation, "If coherence is a test of truth, there is a direct connection with epistemology, for we have reason to believe many of our beliefs cohere with many others, and in that case we have reason to believe many of our beliefs are true." (Davidson, 2000)

 

The Correspondence Theory of Truth

Arguably the more widely-held view of truth (stemming from a broader rationalist tradition in philosophy), philosophers who argue for the correspondence theory hold that there is a world external to our beliefs that is somehow accessible to the human mind. More specifically, correspondence theorists hold that there are a set of "truth-bearing" representations (or propositions) about the world that align to or correspond with reality or states of affairs in the world. A state of affairs just is a particular way the world or reality is. When a proposition aligns to the world, the proposition is said to be true. Truth, on this view, is that correspondence relation.

Take this proposition: "The Seattle Seahawks won Super Bowl 48 in 2014." The proposition is true if in fact the Seahawks did win super Bowl 48 in 2014 (they did) and false if they didn't.

Notice that on this view, propositions about reality are different from beliefs we may have of reality. We believe propositions--I believe that the moon has craters. What follows the "that" is meant to signify the proposition that a person believes. So truth on this view is when the proposition matches reality.

The correspondence theory only lays out the condition for truth in terms of propositions and the way the world actually is. This definition does not involve beliefs that people have. Propositions are true or false regardless of whether anyone believes them. Just think of a proposition as a way the world possibly could be: "The Seahawks won Super Bowl 48" or "The Seahawks lost Super Bowl 48" -- both propositions possibly are true. True propositions are those that correspond to what actually happened.

You'll notice that this definition does not include a belief component. That is, unlike the coherence theory, the correspondence theory describes truth in terms that are independent of beliefs humans may have. This has the distinct advantage of separating truth from the messy business of belief and knowledge but may warrant complaints of being impractical.

Postmodernism

Postmodern thought covers a wide theoretical area but informs modern epistemology particularly when it comes to truth. Postmodern theories of truth are difficult to articulate in strict terms because postmodern theorists tend to eschew hard and fast definitions. But we can provide some insight here. Put in simple terms, postmodernists describe truth not as a relationship outside of the human mind that we can align belief to but as a product of belief. We never access reality because we can never get outside our own beliefs to do so. Our beliefs function as filters that keep reality (if such a thing exists) beyond us. Since we can never access reality, it does no good to describe knowledge or truth in terms of reality because there's nothing we can actually say about it that's meaningful. Truth then is constructed by what we perceive and ultimately believe.

Immanuel Kant

I'm inclined to earmark the foundation of postmodern thought with the work of Immanuel Kant, specifically with his work The Critique of Pure Reason. In my view, Kant was at the gateway of postmodern thought. He wasn't a postmodernist himself but provided the framework for what later developed.  

Kant makes a foundational distinction between the "objects" of subjective experience and the "objects" of "reality." He labels the former phenomena and that latter noumena. The noumena for Kant are things in themselves (ding an sich). These exist outside of and separate from the mind. This is what we might call "reality" or actual states of affairs similar to what we saw in the correspondence theory above. But for Kant, the noumena are entirely unknowable in and of themselves. However, the noumena give rise to the phenomena or are the occasion by which we come to know the phenomena.

The phenomena make up the world we know, the world "for us" (für uns). This is the world of rocks, trees, books, tables, and any other objects we access through the five senses. This is the world of our experience. This world, however, does not exist apart from our experience. It is essentially experiential. Kant expressed this idea as follows: the world as we know it is "phenomenally real but transcendentally ideal." That is objects that we believe exist in the world are a "real" part of our subjective experience but they do not exist apart from that subjective experience and don't transcend the ideas we have. The noumena are "transcendentally real" or they exist in and of themselves but are never experienced directly or even indirectly.

The noumena are given form and shape by what Kant described as categories of the mind and this 'ordering' gives rise to phenomenal objects. This is where it relates to truth: phenomenal objects are not analogues, copies, representations or any such thing of the noumena. The noumena gives rise to the phenomena but in no way resembles them. Scholars have spent countless hours trying to understand Kant on this point since it seems like the mind interacts with the noumena in some way. But Kant does seem to be clear that the mind never experiences the noumena directly and the phenomena in no way represents the noumena.

We can now see the beginnings of postmodern thought. If we understand the noumena as “reality” and the phenomena as the world we experience, we can see that we never get past our experience to reality itself. It's not like a photograph which represents a person and by seeing the photograph we can have some understanding of what the "real person" actually looks like. Rather (to use an admittedly clumsy example) it's like being in love. We can readily have the experience and we know the brain is involved but we have no idea how it works. By experiencing the euphoria of being in love, we learn nothing about how the brain works.

On this view then, what is truth? Abstractly we might say truth is found in the noumena since that's reality. But postmodernists have taken Kant's idea further and argued that since we can't say anything about the noumena, why bother with it at all? Kant didn’t provided a good reason to believe the noumena exists but seems to have asserted its existence because, after all, something was needed to give rise to the phenomena. Postmodernists just get rid of this extra baggage and focus solely on what we experience.

Perspective and Truth

Further, everyone's experience of the world is a bit different--we all have different life experiences, background beliefs, personalities and dispositions, and even genetics that shape our view of the world. This makes it impossible, say the postmodernists to declare an "absolute truth" about much of anything since our view of the world is a product of our individual perspective. Some say that our worldview makes up a set of lenses or a veil through which we interpret everything and we can't remove those lenses. Interpretation and perspective are key ideas in postmodern thought and are contrasted with "simple seeing" or a purely objective view of reality--something postmodernists reject as impossible.

We only have interconnected beliefs and for each individual, that's what truth is. We can see some similarities here to the coherence theory of truth with its web of interconnected and mutually supported beliefs. But where the coherence theory holds that coherence among beliefs gives us reason to hold that what we believe corresponds to some external reality, postmodernists reject that. In postmodernism there is nothing for our beliefs to correspond to or if there is, our beliefs never get beyond the limits of our minds to enable us to make any claims about that reality.

Community Agreement

Postmodernism differs from radical subjectivism (truth is centered only in what an individual experiences) by allowing that there might be "community agreement" for some truth claims. The idea is that two or more people may be able to agree on a particular truth claim and form a shared agreement that a given proposition is true. To be clear, it's not true because they agree it maps or corresponds to reality. But since the group all agree that a given proposition or argument works in some practical way, or has explanatory power (seems to explain some particular thing), or has strong intuitive force for them, they can use this shared agreement to form a knowledge community.

When you think about it, this is how things tend to work. A scientist discovers something she takes to be true and writes a paper explaining why she thinks it's true. Other scientists read her paper, run their own experiments and either validate her claims or are unable to invalidate her claims. These scientists then declare the theory "valid" or "significant" or give it some other stamp of approval. In most cases, this does not mean the theory is immune from falsification or to being disproved--it's not absolute. It just means that the majority of the scientific community that have studied the theory agree that it’s true given what they currently understand. This shared agreement creates a communal "truth" for those scientists. This is what led Richard Rorty to state the oft-quoted phrase, "Truth is what my colleagues will let me get away with."

Practical Concerns

Philosophers are supposed to love wisdom and wisdom is more oriented towards the practical than the theoretical. This article has been largely about a theoretical view of truth so how do we apply it? Most people don't spend a whole lot of time thinking about what truth is but tend to get by in the world without that understanding. That's probably because the world seems to impose itself on us rather than being subject to some theory we might come up with about how it has to operate. We all need food, water and shelter, meaning, friendship, and some purpose that compels us to get out of bed in the morning. This is a kind of practical truth that is not subject to the fluidity of philosophical theory.

Even so, we all contend with truth claims on a daily basis. We have to make decisions about what matters. Maybe you're deeply concerned about politics and what politicians are claiming or what policy should be supported or overturned. Perhaps you care about which athlete should be traded or whether you should eat meat or support the goods produced by a large corporation. You may want to know if God exists and if so, which one. You probably care what your friends or loved ones are saying and whether you can count on them or invest in their relationship. In each of these cases, you will apply a theory of truth whether you realize it or not and so a little reflection on what you think about truth will be important.

Your view of truth will impact how you show up at work and impacts the decisions you make about how to raise your children or deal with a conflict. For example, suppose you're faced with a complex question at work about something you're responsible for. You need to decide whether to ship a product or do more testing. If you're a postmodernist, your worldview may cause you to be more tentative about the conclusions you're drawing about the product's readiness because you understand that your interpretation of the facts you have about the product may be clouded by your own background beliefs. Because of this, you may seek more input or seek more consensus before you move forward. You may find yourself silently scoffing at your boss who makes absolute decisions about the "right" way to move forward because you believe there is no "right" way to do much of anything. There's just each person's interpretation of what is right and whoever has the loudest voice or exerts the most force wins.

An engineer may disagree here. She may argue, as an example, that there is a "right" way to build an airplane and a lot of wrong ways and years of aviation history documents both. Here is an instance where the world imposes itself on us: airplanes built with wings and that follow specific rules of aerodynamics fly and machines that don't follow those "laws" don't. Further most of us would rather fly in airplanes built by engineers that have more of a correspondence view of truth. We want to believe that the engineers that built the plane we're in understand aerodynamics and built a plane that corresponds with the propositions that make up the laws of aerodynamics.

Your view of truth matters. You may be a correspondence theorist when it comes to airplanes but a postmodernist when it comes to ethics or politics. But why hold different views of truth for different aspects of your life? This is where a theory comes in. As you reflect on the problems posed by airplanes and ethics, the readiness of your product to be delivered to consumers and the readiness of your child to be loosed upon the world, about what makes you happy and about your responsibility to your fellow man, you will develop a theory of truth that will help you navigate these situations with more clarity and consistency.


Bibliography

Ackerman, D. F. (1976, December). Plantinga, Proper Names and Propositions. Philosophical Studies, 30, 409-412.

Barrett, W. (1962). Irrational Man. Garden City, New York: Anchor Books.

Brown, C. (1986). What Is a Belief State? Midwest Studies in Philosophy, 10, 357-78.

Brown, C. (1992). Direct and Indirect Belief. Philosophy and Phenomenological Research, 52, 289-316.

Chisholm, R. (1957). Percieving: A Philosophical Study. Ithaca: Cornell University.

Chisholm, R. M. (1989). Theory of Knowledge (3 ed.). Englewood Cliffs: Prentice-Hall, Inc.

Davidson, D. (2000). A Coherence Theory of Truth and Knowledge. In S. Bernecker, & F. Dretske (Eds.), Knowledge: Readings in Contemporary Epistemology (pp. 413-428). Oxford, UK: Oxford University Press.

Dennett, D. C. (1998, August 13). Postmodernism and Truth. Retrieved December 26, 2014, from Tufts: http://ase.tufts.edu/cogstud/dennett/papers/postmod.tru.htm

Frankfurt, H. (2005). On Bullshit. Princeton, NJ: Princeton University Press.

Frankfurt, H. (2006). On Truth. New York: Knopf.

Gettier, E. (2000). Is Justified True Belief Knowledge. In S. Bernecker, & F. Dretske (Eds.), Knowledge: Readings in Contemporary Epistemology. Oxford: Oxford University Press.

James, W. (1907). Pragmatism. Amazon Digital Services, Inc.

Kripke, S. (1980). Naming and Necessity. Cambridge, Massachusettes: Harvard University Press.

Strawson, P. F. (1950). On Referring. Mind, 59 (235), 320-344.

Williams, B. (2004). Truth and Truthfulness. Princeton, New Jersey: Princeton University Press.

A very brief history of western philosophy – Part 1: Plato to Kant

“Philosophy is as old as recorded history and gnarled with the scabs of its ongoing internal conflicts. It is as deep and as tall as the mind can reach. Yet it is also young, even juvenile, in how little of its self-acclaimed potential it has fulfilled, and in how much it stands to learn. It is also old and weary, dying a slow partial death, its heaviest sections withering into unsightly insignificance. ” — an anonymous 20th century philosopher

[About: perennial philosophical questions; brief introduction to classical philosophical themes in truth and epistemology, from Plato to Kant.]
[Length: Approx. 1800 words; 7 minutes to read.]

When people try to figure out what to believe and explain why they do, they don’t spend much time thinking about the fundamental nature of truth, knowledge, or reality. Yet, if you take any belief and keep asking “why?”—keep drilling down for the reasons behind the reasons, you eventually get to the sorts of questions philosophers struggle with. Most of us have neither the time, the interest, or the mental stamina and precision to chase these questions to their ultimate ends. But in certain situations it becomes important to be as certain and as meticulous as one can be in explaining or proving a belief, as in building a bridge, condemning a suspected murderer, deciding which drug to prescribe, or designing the logic of a mission-critical computer model.

One step in assuring a solid conclusion is to start with the most accurate information available. A more subtle determiner of validity involves examining the rules used in the reasoning processes to draw conclusions from the information at hand. That job has traditionally been filled by philosophers. Its not a job most of us would want. Rational beliefs (conclusions that one can give reasons for having) are drawn based on premises or assumptions, so there are basically two problems to work out: The first is how to avoid the infinite regress of challenging the premises (forever asking “why?”) by identifying some fundamental assumptions that all would agree are true. The second is, as mentioned above, how to ensure that the reasoning process guarantees a true conclusion as one steps from premises through intermediate conclusions along the way to the main assertion.


Here are some of the questions classical philosophers have struggled with:

  • What is real? What is true? What can we know (what is the nature and limits of knowledge)?
  • Is the reality of the world different from how we perceive and experience it in our minds? Does physical reality exist apart from the human mind?
  • When something changes or transforms (ages, melts, divides, etc.) it is still essentially the same thing; is its identity preserved?
  • Can consciousness (or ideas, or spirit) exist without the body, outside the physical world? Can pure thought have an impact on physical reality (or vice versa)? Is there anything other than physical reality?
  • Does everything that happens have to have a cause? Is everything that happens predetermined? Is there free will?
  • Can science discover the ultimate nature of realty? Can pure reason (or even intuition) tell us anything about the ultimate nature of reality?

I will reveal the answers to the questions a little later, but for now, it is surprising how many pages have been written and how many lives dedicated to these questions–how many true geniuses have wracked their brains and engaged in prolonged theoretical battles over them–given that we now, finally, have the answers to all of them.

Lets do a bit of a whirlwind tour of some of the central questions, brawls really, of classical western philosophy. Stepping into the ring are well known celebrities such as Plato, Aristotle, and Descartes, and many others whom you may remember from your college philosophy class, including Lock, Hume, and Kant. The contenders are seen as having been members of several schools (clubs or gangs) of thought, including idealism, materialism, empiricism, rationalism, and skepticism (BTW, philosophers are not restricted to being in only one gang).

Plato was an Idealist, who claimed that the world of ideas, for example the ideal nature or essence of a tree or a circle or a color, was more fundamental, more “real,” than physical reality, and that physical reality, a tree for instance, comes into being as an imperfect instance of the ideal. Plotinus, a staunch defender of Plato against Plato’s rivals and misinterpreters, had a more mystical bent, and believed that matter was a manifestation of something deeper and more ethereal. His Platonic philosophy argued that spirit creates the world by stepping from eternity into time and form.

Aristotle was Plato’s student and chief critic. He said “Plato is dear to me, but dearer still is the truth”—ouch! Aristotle was quintessentially practical, none of that invisible eternal spiritual formless essence stuff for him. He worked out some of the basic rules for logic and the scientific method that are still respected today.

Other appreciators of the world of sensation and physical experiences, the Empiricist “I’ll believe it if I see it” team, followed in Aristotle’s footsteps. John Locke, of “tabla rasa” fame, said that the mind starts out without any knowledge and everything one knows is built up from experience through the senses. Bishop George Berkeley one-upped Locke by claiming that things not perceived through the senses can not logically be said to exist at all. His contemporaries assumed that this lead the nonsensical conclusion that the “real” world is an illusion, and, though they could not refute the logic of his theory, rejected it outright.

Scotsman David Hume (who taunted Berkeley by saying that he “often astounds, but rarely can convince”), not to be outdone by an Irishman, brought the Empiricist linage to its skeptical extreme, and, some thought, brought all of philosophy down with it. It would seem that there is virtually no knowledge that we can rationally justify with certainty. We can’t be certain that the cup exists in the just-closed cupboard (we can’t see it now), nor that the sun will continue to rise in the East (just because it always has).

All this was a slap in the face to Idealists, especially Descartes, the “I think therefore I am” guy on the Rationalist team, who pointed out that sensations and experience are famously fallible, so it is pure reason, not the senses, that must form the basis of Truth. For example, what about mathematics? Isn’t that fact that the angles of a triangle add up to 180 degrees True regardless of whether anyone ever saw or even thought about one? Some things are undeniably and eternally true, showing truths that come straight from a pure realm of mind and reason, regardless of the messy, transient world of dirt, blood, and ash.

But alas, reason has its flaws too. Regular old human reasoning, even by the purportedly brilliant, is just not to be trusted, because, heck, reasoning is what lead to all those prior conclusions by other philosophers that each philosopher is arguing against. So we must look for some more formal method of reasoning, such as logic or mathematics. Without going into the details, philosophers since the time of Aristotle have tried so hard, so very hard, to develop systems of logic that would hold up, that would be able to guarantee the truth of some usable knowledge, but each attempt was shot down, its weaknesses and loopholes revealed by the next guy to enter the ring.

So, if we can’t, along with the Empiricists, rely on experience and the seemingly obvious facts presented to our senses, and we can’t, with the Rationalists, rely on pure reason, that equally obvious conviction that C follows incontrovertibly from A and B, where then can we find Truth? The only source left is divine inspiration or intuitively revealed truth. And, all along the historical trail, many were the philosophers who had to lean upon assumptions about God or eternal divine reality to make their case. Long after Plotinus’s mystical pronouncements (he was influenced by Eastern Philosophers), the Bishop Berkley, who had to admit that reality sure did seem to exist even when you are not looking at it, claimed that reality exists because God is always looking at it all. As you might imagine, theories relying on divine metaphysical essence or beings were pretty easy pickin’s for any Rationalist or Empiricists who wanted to take a shot.

Immanuel Kant entered the ring as the Great Referee and Mediator. Kant was a highly respected up-and-coming young philosopher who studied and commented on the eminent thinkers before him. But at the age of 46 he ran into the work of the skeptic David Hume like a brick wall, which, as he put it “awoke me from my dogmatic slumber.” Reeling from Hume’s implications, which threatened to toss all that he had written thus far into irrelevance, Kant entered into a scholarly silence for an entire decade.

He emerged with a key insight: that in order to make sense of the perennial philosophical questions one has to step back and look at the mind itself, to see the mind as a tool; to look at how thought is structured and limited; to be weary of how our concepts influence our conclusions. He is like the first fish who said “hey, I just realized there is this thing I will call Water…” One fascinating question, which I will take up in another essay, is why it took so long, until 1781, for Western Philosophy to generate this perspective. From this starting point Kant developed a body of work, highly regarded to this day, that synthesized earlier views. He affirmed the Empiricist preoccupation with sense experience and the material world, but reconciled this with the Rationalist tradition by noting that experience itself is made possible only through deeper underlying (“a-priori”) mental structures (deeper than concepts or ideas) such as space, time, and mathematical forms like numbers and shapes, that exist regardless of what exists in the material word.

Starting with these insights Kant wrote several classic treatises and thought he had nailed it for once and for all, forcing a “revolution” in thought1 that allowed philosophers to stop squabbling about these fundamental questions and move on to more practical, and even sublime, work. But no. So sorry. Though in some sense he is still seen as reaching some form of pinnacle in Western philosophy, clarifying a host of prior disputes, Kant’s key insight opened up a Pandora’s box that would, over a couple hundred years, bring all of philosophy to its knees, whimpering in despair and confusion, and, in keeping with tradition, fragmented into a profusion of squabbling camps.

(See Essay 8 where I apologize (sort of) for running rough-shod over these brilliant men in the above caricaturizations.)

I know I told you that I would reveal the answers to the list of perennial philosophical questions. And even though at this point in the story it looks like there may be none, I assure you that the answers do exist. But I must sign off here and continue in the next essay, which will include those answers and looks inside the pandoras box that Kant cracked open.

0 comments

Leave a Reply

Your email address will not be published. Required fields are marked *