HPS100 Lecture 02: Absolute Knowledge

Estimated read time: 1:20

Summary

In this lecture, Hakob Barseghyan delves into the concept of absolute knowledge and challenges the idea of proof in science. He questions the possibility of absolute certainty in scientific theories, discussing how historical changes in scientific understanding illustrate that knowledge is not static. Barseghyan examines the differences between mathematical and empirical knowledge, highlighting three major problems—sensations, induction, and theory-ladenness—that prevent us from attaining absolute certainty in empirical sciences. He advocates for fallibilism, arguing that while analytic propositions can be absolutely certain, synthetic propositions cannot, impacting fields like physics, chemistry, and biology. The lecture concludes on an engaging note, teasing a follow-up discussion on how we can compare theories despite the lack of absolute certainty.

Highlights

Hakob Barseghyan opens with skepticism about absolute knowledge in science. 🧐
He presents historical examples like Aristotelian to Newtonian physics. 🤔
Emphasizes the evolving nature of scientific theories. 🌐
Highlights the difference between mathematical certainty and scientific variability. 🧮
Discusses decieving senses using illusions. 👁️‍🗨️
Problem of induction: Just because all swans you’ve seen are white, doesn’t mean all swans are white! 🦢
Theory-ladenness: Your theories shape your observations. 🔬
Encourages fallibilism—an open approach to scientific knowledge. 🤓

Key Takeaways

Scientists can't prove anything beyond a doubt; it's all about questioning the so-called 'proofs'! 🧐
Scientific theories evolve over time; don't get too attached! 🌈
Not all knowledge is created equal; math stands apart! 🧮
Feelings can be deceiving, especially your senses! 👀
Problems in science include sensations, induction, and theory-ladenness. 🧠
Fallibilism rules! Absolute certainty in empirical science? Nah! ❌
Formal science is where the perfect truths reside; empirical science...not so much! ⚖️
Ever wondered why we trust one scientific theory over another? Let's figure it out! 🔍

Overview

In this intellectually stimulating lecture, Hakob Barseghyan questions the notion of absolute knowledge in the realm of science. He challenges the oft-cited claims of 'proof' in scientific contexts, exploring whether science can offer any certainties beyond reasonable doubt. Through historical examples and a philosophical lens, he examines how scientific theories have evolved over centuries, illustrating that knowledge is inherently dynamic.

Barseghyan further distinguishes between different types of knowledge by dissecting the unwavering certainty found in mathematics against the empirical nature of scientific propositions. He outlines three fundamental problems—sensations, induction, and theory-ladenness—which together suggest that absolute certainty in fields like physics and chemistry cannot be attained. His engaging delivery demystifies complex philosophical concepts and relates them to everyday scientific understanding.

Concluding on an explorative note, the lecture paves the way for future discourse on the criteria we use to compare and accept scientific theories despite their inherent uncertainties. Barseghyan's persuasive advocacy for fallibilism underscores a more open-ended and questioning approach to empirical knowledge, inviting audiences to navigate this scientific landscape with intellectual rigor and curiosity.

Chapters

00:00 - 00:30: Introduction to Absolute Knowledge The chapter begins with a focus on the narrator's irritation when encountering certain passages, suggesting an exploration of the concept of 'absolute knowledge'. The introduction might delve into philosophical discussions about knowledge, perception, and perhaps the frustration with misunderstanding or lack of clarity in learning. It sets the stage for a broader examination of the themes surrounding human pursuit of knowledge.
00:30 - 01:00: Philosophical Questioning in Science The chapter explores the idea of proof and certainty in science from a philosophical perspective. It questions common phrases like 'Scientists have proven that...' and 'Science has established beyond any reasonable doubt that...' It challenges the reader to consider whether science can offer true proof or absolute certainty.
01:00 - 02:00: Scientific Theories Through Time In this chapter, the timeline of scientific theories is examined, highlighting how accepted theories have evolved over time. Present-day accepted theories, those from 250 years ago, and theories from 500 years ago are placed on a timeline. The chapter poses a question regarding the evolution of scientific theories over time, using the historical shift from Aristotelian physics to Newtonian physics as an illustrative example.
02:00 - 03:00: Change in Theories Across Disciplines The chapter discusses the concept of change in theories across various disciplines over time. It begins by confirming that this evolution of theories is not unique to physics but is a common theme across all fields. Highlighting examples from astronomy, it traces the progression from geocentric models of the universe, believed 500 years ago, to Keplerian elliptical astronomy 250 years ago, and on to modern cosmology with theories such as the Big Bang.
03:00 - 04:00: Questioning Certainty in Knowledge The chapter explores the theme of certainty in knowledge, questioning whether any element of our understanding of the world is immune to change. It discusses the historical fact that theories evolve over time, prompting an investigation into whether absolute knowledge exists. The speaker intends to provide three straightforward examples to further explore this query.
04:00 - 05:00: Example from Mathematics The chapter titled 'Example from Mathematics' begins with an introduction to a basic mathematical equation: 1+2=3. The discussion then pivots to the concept of understanding and justifying mathematical truths, beyond mere acceptance of them as taught, questioning the fundamental reasons why this proposition is true.
05:00 - 06:00: Observational Justification in Mathematics The chapter explores the concept of observational justification in mathematics, beginning with a scenario illustrating how ancient people might have noticed basic arithmetic through observation. As they observed that adding different fruits together resulted in consistent sums, this served as an early form of justifying mathematical concepts through direct observation.
06:00 - 07:00: Truth in Mathematical Propositions The chapter discusses the concept of truth in mathematical propositions, using the analogy of counting bananas to illustrate how individual experiences can lead to generalizing mathematical truths. The speaker humorously questions whether this method, such as counting cockroaches, justifies truth in mathematics, and asks the audience if they agree with this view.
07:00 - 08:00: Synthetic vs. Analytic Propositions The chapter titled 'Synthetic vs. Analytic Propositions' opens with a rhetorical question aimed at the audience, questioning whether current mathematical processes are valid. The lecture seems to challenge the traditional understanding of mathematics, suggesting that it does not necessarily rely on experimental validation. A student responds by suggesting that mathematics relies on set theory to define fundamental concepts like numbers, equality, and addition, implying that these definitions inherently hold truth. The discussion seems to pivot around whether mathematical truths need empirical evidence or are self-sufficient through logical definition and reasoning.
08:00 - 09:00: Examples of Synthetic Propositions The chapter titled 'Examples of Synthetic Propositions' discusses the nature of mathematical propositions and the role of definitions. Hakob emphasizes that mathematical truths are derived not from empirical observations, like seeing apples or cherries, but from starting with definitions. For example, the understanding of the number '2' is based on the definition of having one item and another item. This demonstrates how mathematical concepts are often true by definition rather than through experimental validation.
09:00 - 10:00: Propositions in Empirical Sciences The chapter discusses the foundational concepts of understanding numbers and propositions in empirical sciences. The narrator elucidates that understanding numerals, such as "2" and "3", doesn't necessarily require physical counting or observation but is based on definitions and an understanding of aggregating singular instances. This understanding is applicable universally, regardless of tangible examples like apples or cherries, emphasizing a conceptual grasp over empirical observation.
10:00 - 15:00: Problem of Sensations The chapter "Problem of Sensations" explores the concept of numbers and language. It discusses how the interpretation of words and numbers is tied to their established definitions in English. For example, attempting to redefine 'three' as 'five' by simply changing language does not alter the core concept of 'three,' which is essentially the sum of 'one and one and one.' This illustrates the limitations of altering fundamental concepts through language changes.
15:00 - 17:00: Issue with Induction The chapter titled 'Issue with Induction' explores the fundamental nature of mathematical reasoning. It describes how conclusions are reached not through empirical experience but through deductive logic based on definitions. The speaker asserts that mathematics generally follows a pattern where definitions (such as those of numbers and operations) serve as fundamental building blocks. From these definitions, theorems are logically deduced, which further aid the mathematical discourse. The chapter emphasizes the logical and non-experiential nature of mathematics.
17:00 - 19:00: Theory-Ladenness of Observations The chapter discusses the concept of theory-ladenness in observations. It begins by talking about the foundational principle in mathematics where theorems can be used to deduce further theorems, illustrating a chain of reasoning inherent to mathematical practice. The discussion shifts to a more challenging case, exemplified by the statement 'all swans are white'. This example is used to question how certain observations or statements can be justified, questioning whether such statements can be deduced merely from definitions or require empirical justification.
19:00 - 21:00: Summary and Fallibilism The chapter discusses the concept of fallibilism in relation to the color of swans. It suggests that just because we define swans as white based on previous observations, it doesn't necessarily mean all swans are white. The conclusion is that propositions should be made based on real observations and empirical evidence, rather than preconceived definitions.
21:00 - 21:30: Final Thoughts and Next Discussion The chapter 'Final Thoughts and Next Discussion' discusses the process of inductive reasoning using the analogy of observing swans. Initially, an observation of a single white swan leads to a specific proposition ('swan A is white'). With further observations (like finding a second or third white swan), one can generalize a broader inductive conclusion (e.g., 'all swans are white'). The narrative suggests a logical progression from specific observations to general conclusions, emphasizing the straightforward nature of this inductive process. The chapter ends by noting a distinction to be addressed in future discussions, hinting at nuances in the method that will be explored.

HPS100 Lecture 02: Absolute Knowledge Transcription

00:00 - 00:30 If there is one thing that really annoys me, it is when you come across a passage
00:30 - 01:00 - and this is an often recurring passage - when people say "Scientists have proven that ..." or "Science has established beyond any reasonable doubt that..." As a philosopher, you cannot help but have a question: "Is there such a thing as proof in science?" "Can science actually prove anything beyond any reasonable doubt?"
01:00 - 01:30 Let's take a timeline now. We have a timeline. And now here we have theories we accept nowadays. These would be theories accepted 250 years ago. And those would be theories accepted 500 years ago. I'm going to move all of them to one side here. And we have a very simple question for you: We've seen last time that scientific theories change through time. We have to take this historical fact as our starting point. One example we discussed was the transition from Aristotelian physics to Newtonian physics
01:30 - 02:00 to the contemporary general relativity. But the idea that theories change through time is not only specific to physics - this is the case for every discipline. You can take astronomy for instance: Five hundred years ago they believe that the Earth was in the center of the universe. Then 250 years ago there was Keplerian astronomy (elliptical astronomy). And nowadays we have cosmology with big bang theory and all sorts of things.
02:00 - 02:30 It's a historical fact that theories change through time. Now the question is if that is really the case then in this mosaic is there anything unchangeable? Is there any element that doesn't change - that is immune to change? Another way of putting the same question would be: Can we know anything for certain? Can we know anything with absolute certainty? Is there such a thing as absolute knowledge? I'm going to give you three examples that are very straightforward
02:30 - 03:00 We will start with a very basic example from mathematics. This one: 1+2=3. Now, tell me: How do we know this? And I'm not asking how you personally acquire this knowledge. You probably were three or four when your mother or your father told you this. That's not what I'm asking. What I'm asking is: how do we justify this? In other words, what makes this proposition true?
03:00 - 03:30 I'm going to give you one scenario and you tell me if you agree with that. How about I tell you that what I think what happened is long long time ago people started to notice that when you take one apple and then add to this apple two more apples, then you count them and you get 1, 2, and 3. Then you come across a cherry put together two more cherries and you get the same result.
03:30 - 04:00 And then the same happens with bananas: one banana, two bananas, and you get three bananas. An then you generalize this from your basic experience for individual cases, you generalize and you arrive at the general proposition the one plus two equals three at all times. I've never tried this with cockroaches: you will end up with thousands of them. Now tell me is this how we really justify truth in mathematics? How many of you agree that this is exactly how it works in mathematics? 1,2, 3, 4 ... OK!
04:00 - 04:30 How many of you believe that this is utter nonsense - that this is not how things happen in mathematics? OK! Very good! I'm with you! So one of you tell me why you think this is not the case! (A student): "I think mathematicians use set theory to find numbers and equality and addition and then give the definitions of these terms and this ends up being true." (Hakob) In simple language, what you are saying is that you don't really need experiments
04:30 - 05:00 and observations in order to get to this? (Student) "Yeah!" (Hakob) That is exactly the case! The thing is you don't really need to come across apples or cherries or any other objects in order to show that this is actually the case. What happens in mathematics is that we really start with definitions. When I say "2" - the concept of "2" - what is it that you have in mind? It's one and another one. This is true by definition!
05:00 - 05:30 You don't really have to do any calculation to come up with this result. Right? So essentially when you say "2" you mean one of something and another one of something - together. That's your definition of "2" and that's what you mean by "2". The same applies to "3". It's the same same idea: it's one another one and another one. That's true by definition. You don't really have to count or observe things. This would hold even if you didn't have any apples, even if you don't have any cherries.
05:30 - 06:00 You don't really need that! Is this clear? This is the meaning of the words the way we use them Although you can say "you know what, I'm going to create my own language in which "three" means one plus one plus one plus one plus one." (A synonym for what we call "five" in English.) Well, you can certainly do that, but that would be just another language. You wouldn't change the concept of "3". The concept of “3" is basically one, another one and another one.
06:00 - 06:30 That's true by definition. Now if you put these two together, you arrive at this conclusion deductively, logically. You don't need any experience for this. Can we all agree on this? Very good! Now it's safe to say that the whole mathematics follows the same template. So you get some basic definitions - definitions of numbers, definitions of mathematical operations. From these definitions you deduce certain theorems and then you can take this theorem and that definition
06:30 - 07:00 you deduce another theorem and then you can take this theorem and that theorem to deduce yet another theorem. The whole mathematics is like this. Anything in mathematics is is going to be along these lines, OK? Very good! Now let's take another case - a more difficult one. All swans are white. How do we know this? How can we justify this? Is this something that follows from definitions? You think you can deduce this from the definition of swan?
07:00 - 07:30 Are all swans white because that's what the word "swan" means? Probably not! So the intuitive answer would be that we have to really go out there, find real swans, observe them, fix the results and then and only then see whether this proposition makes any sense! OK, so let's do it. Let's go find a swan. We find a swan, we see that it's white and we fix the result in a proposition. This will be a single of proposition that describes our observation.
07:30 - 08:00 So you see a swan and you say "swan A is white." When you find another one, you fix this in another proposition. If it's your really lucky day, you find the third swan. From all these cases you generalize by induction and arrive at a general conclusion. Makes sense? Sounds along the lines of what you have in mind? Very good! This is very straightforward! Now pay attention to the difference
08:00 - 08:30 between this example and the example before: the mathematical example you didn't need any observation, you didn't need any experience but here you need that! There's no way to know this unless you go out there and observe. To say this is the same as to say that our theory that all swans are white - this is a theory, a general proposition - it is based on your experience -
08:30 - 09:00 individual experience of white swans. Basically when we say experience will mean observations and experiments. You know the difference? (Student) "Observations would be naturally occurring and experiments would be something that is fabricated in order to get result." (Hakob) Thank you! The idea is experiment assumes an artificial set up and in observation you just sit there and look. In an experiment you create
09:00 - 09:30 an artificial situation. That's the difference! So here you have a basic template: you have individual observations, then you fix these observations in single propositions and then you arrive inductively at the general conclusion. This is what you have. Now let's take another example: the law of gravity. For those of you who are not really familiar with physics, that's totally fine. But you have to understand the basic idea.
09:30 - 10:00 You don't have to remember the equation. OK? Don't be afraid! This is the idea: the law says that if you have two objects, Mass m1 and Mass m2, and r is the distance between the two then there is always a force - a force of attraction between the two objects and this force is equal to the product of the masses and inversely proportional to the square of the distance. So basically it tells us that there is a force of attraction between any two objects in the universe
10:00 - 10:30 no matter how far removed form each other they might be. This force is greater when the masses are greater and this force is smaller when the distance is greater. This is what the law tells us. This is very straightforward, right? Not very difficult. You know the law - this is the high school stuff - now tell me how do we know this? How do we justify this? You have two options: either this is similar to the case of mathematics when you can
10:30 - 11:00 dedice this from some sort of definitions - I don't know what kind of definitions they might be. Or this is similar to the case of white swans when you actually have to go out there and observe things. How many of you think that this is going to be a very similar to the case of mathematics? 1, 2, 3 ... OK! Very few. How many of you think that this is going to be very similar to the case of white swans - and you are correct !
11:00 - 11:30 This is not something that you can deduce from a definition of physical object or distance or force. You have to actually go out there and observe. There is only one way. That seems to be the intuitive answer and that's exactly what we seem to be doing. So what do we do? We go out observe falling apples and we notice that the law holds for the earth and the falling apple. And you fix this. You have a single proposition describing experience. Then you observe that the same
11:30 - 12:00 law holds for folding water and the earth and you fix this. Then you observed that the same law holds for any two planets and you fix the result. Based on these results you finally use induction and you arrive at this conclusion inductively. Is this clear? Very good! So we basically had three cases and as you can see these cases belong to two different categories:
12:00 - 12:30 The cases of white swan and the law of gravity are very similar but the mathematical case was somewhat different. There is a reason why they appear different because they are essentially different. Mathematical propositions are analytic propositions. And these ones are called synthetic propositons. I'm going to explain what they are. Analytic propositions are those deducible from definitions because they are true by definition.
12:30 - 13:00 They can never contradict any observation, no matter what happens they can never contradict - you can never come across a case when this is actually not the case! If you ever happened to meet a bachelor that just happens to be married, is it possible? You may just say well I could be using the language differently! Maybe what you mean by bachelor is not what people commonly mean by bachelor therefore if you mean by bachelor what we commonly mean then it's impossible to come across a
13:00 - 13:30 married bachelor - that would be a contradiction in terms! So you can never have contradiction with experiments and observations. And another way of putting - this is a just equivalent formulations - in a fancy philosophical language: analytic propositions hold in all possible worlds, meaning that you cannot possibly conceive a world where this is not the case. One plus two equals three in all possible worlds. You cannot possibly
13:30 - 14:00 conceive a world that is not the case. Now we take a look at synthetic propositions. They are not deducible from definitions. They can contradict the results of experiments and observations. The opposite is conceivable. I will give you a few examples here. All propositions of formal sciences will be here, mathematics and logic, everything, this, this, a few more; they're all are going to be here.
14:00 - 14:30 Even if you have never taken any basic logic believe me it's all going be here on this side. On the right side you are going to have things like the law natural selection for instance - the idea that species evolve - they evolve according to certain logic. Why is this the case? Because you could easily conceive a world where this is not the case. You can easily imagine a world in which species are created at one point of time and they never evolve. Can you imagine this? There's no contradiction in this.
14:30 - 15:00 I'm not saying this is true but the rule of thumb is if you can conceive of the opposite, then that's a synthetic proposition. OK? That's the idea. One more example from economics: diminishing returns. Any one knows what the law of diminishing returns is? (Student) "Diminishing returns refers to the gains you receive from doing something more than once is reduced each time you continue to do the thing."
15:00 - 15:30 (Hakob) Very good. That's basically the idea. If you only produced let's say 100 iPhones your return per unit is going to be let's say $10 or $100, but if you keep producing them - if you increase the quantity - then return per unit is going to decrease and essentially you're going to saturate the market to a degree when you don't get any returns. That's basically the idea. Now what unites all these laws they all come from empirical science: physics, chemistry, biology, psychology,
15:30 - 16:00 sociology, political science, cultural studies. These are all empirical sciences. Their propositions come from experiences and none of them is true by definition. These are all hypotheses about the world. Let's get some more examples. Can we easily conceive of a world in which the laws of physics are such that the floating mountains are possible? It is not the case in our world you realize that's not really possible but you can easily imagine such a world. What about this one:
16:00 - 16:30 can we easily imagine a world with an alternative set of biological laws where biology says that you can have fire-breathing evil dragons? Can we easily imagine a world with a different geography when continental Westeros actually exists? This is what makes physics, chemistry, biology, and all other empirical sciences synthetic propositions because the alternative or the opposite is conceivable. OK?
16:30 - 17:00 That's the reason why they cannot be true by definition. You have to go out there and observe. We just happen to inhabit such a world where we cannot have fire-breathing dragons but they are conceivable - not possible according to our laws but conceivable: meaning that you can imagine such a world! You understand what I'm saying? Very good! Now this was a basic question and since we have two types of propositions - synthetic and analytic, we have to differentiate two different questions.
17:00 - 17:30 We have to specify whether there can be absolutely certain analytic propositions? A similar question can be formulated regarding synthetic propositions: can we have absolute certainty when it comes to synthetic propositions? Let's start with the first one. What do you think about the first one? Can be have absolutely certainty when it comes to definitions? The answer is yes. It is a yes because these things are just true by definition. If you stick to our definitions then
17:30 - 18:00 nothing whatsoever can challenge this. You can never come across a single observational result that contradicts it. Therefore this will stay forever. The whole mathematics is like this and the whole formal sciences is like this and therefore the answer is yes. Unfortunately, we're not in a position to say the same thing concerning synthetic propositions and this is where the situation gets really complicated. To answer this question we have to find out how these propositions
18:00 - 18:30 are justified. The intuitive answer has to do something with experience. We've already covered this - just like in the cases of of swans or law of gravity - has to do with experience. And if we think of a general template, this is we have: basically what we do with synthetic propositions is we find an object and then we notice that this object - let's call it P1 - has certain
18:30 - 19:00 property whatever the property is. P1 has a property of Q. When you try to find something similar, say, another abstract object, and you notice that it also has property Q. The same applies to yet a third object. From these cases you generalize and you find that all Ps have property Q. So this is the general template I think we can all say and all agree that general synthetic propositions must be somehow based on experience.
19:00 - 19:30 There is no way to know these things just by definition. Can we all agree on this? You have to really go out there and observe things. Very good! But this is the reason why they cannot be absolutely certain. Why? Because of three problems. Today I'm going to give you three reasons - three problems or three arguments - why you can never have
19:30 - 20:00 absolutely certain synthetic propositions. Problem No. 1: Sensations. Here we have the human mind - your thoughts, your emotions, your feelings and your theories ... everything in your mind. Here we have the external world. So here we would have swans as they exist in reality. Another way of saying this; swans as exist objectively. These are the swans as they exist out there.
20:00 - 20:30 And here we have swans as they are perceived by a human being. These would be my sensations of swans - swans as I see them. Understand the difference? Between the two? One thing is how things exist in reality and it's another thing how you perceive them: they may or may not coincide but at this stage we don't know. So we have to separate the two. You see what I'm doing? Very good!
20:30 - 21:00 Now, this is the basic question: How can we make sure that the swans as we see them are exactly as they are in reality? This is the question. How can we ever find out whether our perceptions and sensations provide us with the exact image of things as they exist in reality? Do you understand the question here? What I am going to do in the beginning, I'm going to tell you that we have at least some reason to suspect that our senses are at least questionable!
21:00 - 21:30 How about famous illusions such as this one here? Do you see a bubble in the center of the image? Do you actually see the bubble? But you do realize that there is no bubble? Can you all see this famous illusion? What about these lines getting narrower or wider? Do they? You understand that in reality they don't change! And this is my favorite one: Which of the two squares is darker?
21:30 - 22:00 A or B? Which one is darker? A appears to be darker for sure! Right? Can we all agree that at least it appears to be darker than B? You all know the answer. It's exactly the same color! What all these examples tell us is that senses are at least questionable. This is not the major argument why we think that senses are questionable - this is just a basic idea to begin with.
22:00 - 22:30 This is the major problem: Suppose I see that there's a cup of coffee in front of me, I have a sensation of a cup of coffee: it's here in the brain. Now how can I make sure that there is really a cup of coffee in front of me? Any ideas? You have a visual sensation and now you want to make sure that the it is not deceiving you. What would you do?
22:30 - 23:00 (Student) "You could taste it or touch the coffee to make sure that your other senses are in agreement with their sight!" (Hakob) You name is? (Student) "My name is Anthony." (Hakob) Very good, Anthony! So the in intuitive answer seems to be "Well, just let's go out there smell it tasted it and touch it." That's exactly what we are going to do. I can try to confirm my visual sensation here - but my sensation of smell ... you all know that what makes coffee coffee is really not the taste but the smell, you know that right? I had a girlfriend used to say it's not even smell but the sound of
23:00 - 23:30 the morning coffee! It might be a little extreme but we can all agree that smell is at least as important as taste, if not more mportant! Then what do you do? You taste it - finally you have to get there and you have to drink it. So what do we do here? By doing this we basically try to confirm our visual sensation by the sensations of smell and taste. Does it solve the problem?
23:30 - 24:00 (Student) "The sense of smell is just as easily confused as the visual sense of things. So we can't really be sure about other senses about whether or not the cup of coffee is there because technically we can't trust other senses either?" (Hakob) That's exactly the point! That's exactly the point! The only thing that we have been able to establish is that there is some sort of coherence between our sensations. That's basically it. It doesn't tell us anything about the thing in itself. In order to be in a position to say well
24:00 - 24:30 you know what I've confirmed my visual sensation and I can be absolutely sure that it's not deceiving me what I have to do is I have to be able to take my sensory mechanism - all my senses - and put them aside and look at the thing as it exists in reality without actually looking at it, which is quite impossible I guess! Even if it turns out that human beings have the so-called "sixth sense" or the "26th sense",
24:30 - 25:00 Will it change anything? OK! You would be in a position to confirm you visual sensation with seven more other sensations but essentially you're going to remain in the world of your sensations no matter what do you see, what you taste or what you hear it's all going to be a sensation. You can ask one of your friends to confirm it for you, but will this solve the problem? It wouldn't solve the problem. Becuase, how do you know what he tells you? You perceive it through your sensations. So essentially the only thing
25:00 - 25:30 to deal with is a complex of sensations and that's what you have - it's all sensations. Makes sense? You all agree that there is a problem here? Very good! So essentially this is where we stand: in order to be in a position to say that the swans are really white, you have to be absolutely sure that your senses conveyed the exact picture of the Swans as they really are; but in reality this
25:30 - 26:00 cannot be guaranteed! It's very unfortunate: they cannot be guaranteed! The only thing that can be guaranteed is that I see swan A is white and swan B is white. The only thing can be guaranteed is that you see a white swan in front of you - not that it is really white! You might be thinking that "you are really picky here!" But that is what philosophy is all about! You have to take ideas to their extremes and see what happens.
26:00 - 26:30 The question here was not whether you can trust your senses. Of course for all practical purposes you have to trust your senses, otherwise you wouldn't survive. You have to do that. But the question is: Are we in a position to say that we have a sensation which is an absolutely correct image of the thing as it really is this is? This is the question, becuase in order to be absolutely sure that this swan is exactly white and not whitish, what you have to do - you have to be in a
26:30 - 27:00 position to say that you can question my sensation of smell and taste - I may or may not have a good taste - that's what you question it but you cannot question my visual sensation. When I see a white swan, it is really white. You have to be in a position to say this. But are you in a position to say this? No! This is what's really unfortunate. We are never in a position to be absolutely sure that our senses are not deceiving us. Is this clear?
27:00 - 27:30 (Student) "Is the unreliability of senses limited to human sensations? Let's say I don't know if this is scientifically possible but let's say there was a machine that mechanically analyzed whatever it views? Will that sensation also be unreliable?" (Hakob) Do you understand the question? Anyone wants to reply? (Student) "In order for you to perceive what machine is telling you, you have to use your
27:30 - 28:00 own sensations to do that. So your sensations can still be lying to you." (Hakob) Everyone agrees with that? You see the problem is more general even if we come to a point when we as human beings become really cyborgs - you know what cyborgs are, right? They are half human half machine - even when we come to that point, when we come to the point where we start to replace our senses with some alternative vision mechanism or other senses, that wouldn't really solve the problem because essentially they would remain senses - they would remain sensations.
28:00 - 28:30 Make sense? (Student) "Can I put it in terms of an example: let's say we have the cup of coffee and instead of looking at it or smelling it or tasting it, I put it inside a machine that will analyze the chemical makeup of whatever is in there and it will tell me this is made of water and coffee bean powder?" (Hakob) That's very good point. Any replies? (Student) "I think that if it is a man-made machine then the answer
28:30 - 29:00 would be no, because we don't know what to look for besides what we already see. And I don't see how we can develop a machine that looks for more than what we already know." (Hakob) I completely agree. The problem here is that in order to first develop that machine you need some sort of a theory. You have to have very elaborate chemistry to begin with and maybe also some fundamental physics in order to build a machine. And at this stage we don't have anything.
29:00 - 29:30 Basically every sensation is questionable because no matter what sort of a mechanism, what sort of a contraption will come up with, its always going to be based in a circular fashion on some other sensations, and that's the point! That's basically it! (Student) "I was just wondering about colour and how we see colour. You said we can't say 'all swans are white' but colours are also about definitions as when we say something is red, it's because it is defined as being red.
29:30 - 30:00 Although everyone see colour a little different and for myself its not absolute if there is no definition." (Hakob) I see your point. Anyone wants to reply? (Student) "In order to answer your question, I think we should probably first consider what is is a swan? I think maybe the definition itself is a concept of sensation." (Hakob) What you're trying to get here is a more fundamental problem which I was not
30:00 - 30:30 going to touch upon in this course ... maybe only slightly: it is the idea that you know to have even very basic individual propositions describing the experience you need some sort of language and chances are you cannot really come up with the language unless you had some previous experiences before. That is true. But for the sake of argument suppose that this problem doesn't exist; that we know what whiteness is, that we have a strict definition of what would it mean
30:30 - 31:00 for something out there to be white - going back to the previous point - suppose we had a clear-cut physical definition - let's say electromagnetic radiation with such-and-such wavelength - suppose we have that definition, came up with it somehow, now forget about the problem of definitions for a second, let's be generous for the sake of argument suppose again we have a clear-cut idea what swans are and what whiteness is, are we really in a position
31:00 - 31:30 to say that whenever I perceive whiteness, does it mean that the thing out there is really white? This is the question. You see the problem here? So regardless of how we came up with the idea of whiteness or swans we would still have that problem. Make sense? (Student) "Senses are not restricted to human beings.
31:30 - 32:00 Animals also have senses. The reason we have a wrong perception is because our intellect and we can perceive things differently the way we see. Animals don't have that so their senses would have to be perfect." (Hakob) Some people say you can trust your senses because you know that human beings are a result of long lasting evolutionary process. Basically all our senses are adjusted to our environment such that we perceive it more or less
32:00 - 32:30 correctly because if we didn't we wouldn't survive. How do you like this argument? (Student) "The problem with the argument that our senses are reliable because they are products of evolution is that evolution is a theory that we only know based on evidence that we were able to attain through our senses. So it's circular." (Hakob) It certainly is. You see any time we use any of the accepted theories to solve any of these problems, we end up in a circle. Because at this stage we don't
32:30 - 33:00 know anything - you can't use any of your accepted scientific theories - that would be begging the question, there is no easy way out: you have to try and establish the trustworthiness of your senses regardless of all the achievements of evolutionary biology, neural physiology, physiology of sensations, and psychology. You can't use any of that because those will be theories. You can only trust your evolutionary theory if you rely on thousands and thousands
33:00 - 33:30 of observations. So you cannot assume that theory from the very beginning to prove the trustworthiness of the observations. You would end up in a vicious circle. OK? So that's why we have a problem here. Now is this clear that we have a problem here? Just to clarify this is the last point I'm trying to make: the point here is not that your senses are deceiving you. The thing is that we are not in
33:30 - 34:00 a position to know whether they are deceiving us or whether they are not! That is the point! If we knew that under such-and-such circumstances senses are not trustworthy that would solve the problem. You say "well on such and such circumstances they're not trustworthy", you mean that under different circumstances they are trustworthy. So that would solve the problem but the very issue is in the absence of this touch stone. We are not in a position to know when they can be trusted and
34:00 - 34:30 when they cannot be trusted. That's the whole thing. That's the whole problem. Problem No. 2: induction. Let's forget about the problem of sensations. Suppose you are in a position to trust your senses at all times - you have never heard about the problem sensations - and anything you perceive provides you a hundred percent correct picture of reality, for the sake of argument. OK? There is another problem which is as fatal as the first one! This is your scheme. How can we possibly arrive at
34:30 - 35:00 this general conclusion? Pay attention that it says "all“ - it doesn't say this or that, says all objects of the same class. How can we have possibly arrived at this conclusion if experience provides us only with singular propositions? You see the difference here? Whatever you perceive, you perceive only
35:00 - 35:30 individual things, individual objects. You never see all the swans or all the planets or all the things. It's always individual: it's always this or that. Now the transition itself is what we call induction. This thing here, that allows us to transition to general theories from individual experience. How can we ever do this? How is this possible? Let's take a timeline. This is the present; that is the future; that is the past.
35:30 - 36:00 We have the swans observed up until present. We fix the results of observations in individual propositions describing experience and from this we generalize and arrive at the proposition that all swans are white. Now the thing is the moment you accept this, what you really accept is not that only these three swans are white. That's not what you accept. The moment you say that all swans are white basically what you're saying is that all swans that
36:00 - 36:30 have been born until present are white, and the swans that will be born in the future are also going to be white. That's what general propositions are all about. Because they refer to all elements within the class. So in this case it means that in the future whatever swan I observe is going to be white. Do you think there is problem here? No matter how many millions of white swans you end up
36:30 - 37:00 observing there is always a chance of coming across a black one. There is always the possibility. There is no guarantee that you have observed each and every one of them even if you have managed to cover each and every one of the current swans. How do you know about the future swans? This is where the problem is. So the moment you come across a black swan you arrive at a contradiction which contradicts your general proposition and as a result of that it fails and it's no longer absolutely certain.
37:00 - 37:30 You see the problem? OK. So this is the basic template of the problem. The problem is that our experience is always limited even when you rely not only on your personal experience but also the experience of your peers and experiences in previous generations, it is still limited. You are never in a position to observe every combination of planets in the universe. That's impossible that's a task is if you want to establish a law of gravity.
37:30 - 38:00 It's impossible. You cannot put together all possible combinations. You cannot test everything. You cannot observe everything. We have to accept this as an unfortunate fact of life. Experience is always limited. And since it is limited, inductive generalizations are also inevitably fallible. You understand the problem? Any deas how this can be solved or bypassed?
38:00 - 38:30 (Student) "Why don't we solve the problem by just changing the theory to 'most' Ps have the property Q?" (Hakob) That wouldn't give us absolute knowledge. That would limit our knowledge to current cases, cases observed so far only. That certainly solve the problem, but that wouldn't give us what we are looking for: absolute knowledge, right? That would be a step back. In as sense that would be a defeat. (Student) "If we did something such as finding the best the way to do something, or optimization.
38:30 - 39:00 Say you are trying to get to a Starbucks store, the best way is always to plan out this route, right? So you're not going be like, "hey I'm gonna roll a dice and try to see, like, how many steps they should take this way and that way." (Hakob) I guess your point is: there is a reason why everyday we choose the same path. And the reason is that our previous experience tells us that that is the shortest road to the Starbucks. That is correct except that is not
39:00 - 39:30 absolutely correct. It is not absolutely correct because it is based on an assumption that nothing is going to happen to the road in the future. Are you going to have the same arrangement of streets and the same arrangement of houses? Right? How many assumptions you are making? How many of those assumptions are safe? Not really. How about I say that you know what I don't really have to observe each and every swan because I know for a fact the future is going to be exactly like the present and the past
39:30 - 40:00 because there is a certain uniformity nature. Let's say I want to establish the proposition that all human beings are mortal. We have billions and billions of cases of human beings dying. That would be my limited experience and you say well you know what you don't really have to wait until the whole humanity becomes extinct - you don't really have to wait for that - because you know that the
40:00 - 40:30 future is similar to the present so there is some sort of a continuity in the internal mechanism that makes human beings mortal. You see what I'm trying to say? If this were the case - if this were true - I would be in a position to say well you know what you don't need to make an infinite number of observations. A limited number of observations would be enough. You observe, say, 10 human beings, thousand human beings or millions of human beings and that would be enough to say that any being that will be born in a future is also going to be mortal.
40:30 - 41:00 Is it a good solution? (Student) "The problem is that we have no guarantee that the future will be like the past so how can we be sure that just because the swans were white yesterday and today that they'll be white tomorrow?" (Hakob) The solution that I proposed about future being similar to the past ? It doesn't really do the job because how do I know it's similar? Or you can say we've been here in this world for thousands and thousands of years and we can say well we've observed that the
41:00 - 41:30 future so far has been similar to the past and from this I deduce use the future is always going to be similar to the past. What am I doing here? I'm using induction. I have to use induction to arrive at a principle that future is always going to be similar to the past and then use that principle to save induction and then use induction to save that principle and then save that
41:30 - 42:00 principle in order to use induction. It's an endless circle. You understand? So the problem doesn't have a solution, and its accepted by a contemporary philosophy of science that this issue has no solution - the problem of induction. So essentially it's not solvable. But for all practical needs and purposes you have to trust your previous experiences. There is nothing else you can do. But you have to concede that it doesn't guarantee absolute knowledge!
42:00 - 42:30 This is the only thing that you have to concede. Not that you have to stop making inductive generalizations, but what we have to concede is that there are no guarantees that this is a hundred percent safe. You see the point I'm trying to make here? (Student) "I was just wondering if induction works backwards? Can we go from what we observe in the present to what happened in the past?" (Hakob) Let's say you want to calculate specific configuration of planets in our solar system, 4000 BC.
42:30 - 43:00 What do you do? You take your current theories as your starting point. You take your current observations of the current positions and you calculate backwards. And this is what we do in order to use that knowledge in historical chronology. Let's say you came across a manuscript which says that such and such events happened a year after a solar eclipse and three years before the next solar eclipse. So what do you do? You take your astronomical knowledge as a
43:00 - 43:30 starting point then the question becomes when exactly the sort of a conjunction happened in the past? So that's that's exactly what we do here. OK. There are no more questions about induction, then let's agree that there is an issue. Can we all agree that there is a problem that makes the possibility of absolute knowledge questionable? If this is the case, then we move on to Problem No. 3 And this is my favorite of all: theory-ladenness. What do we see here?
43:30 - 44:00 Those of you who know the answer please be quiet for now ... those of you who are professional in the field. What do we see here? I guess a layman would say "well I only see some fuzzy and grayish picture." But an educated person will see a synapse. What an educated person would say
44:00 - 44:30 "Well, it's straightforward. What you have here is the end of an axon here, and here you have the synoptic knob and here you see the synaptic vesicles, here you have a synaptic cleft that separates the axon from the dendrites." This is what you see if you're educated in neural physiology, meaning if you are told what to see. I guess you can guess where I'm heading to with this.
44:30 - 45:00 Let's zoom out: an experienced person with experiences of something along this line, what happens here is that you get some neurotransmitters that cross the thin gap between this axon here and this dendrite here. - the so-called synaptic cleft - and are picked up by neuro-receivers in the dendrite's membrane here in this in this region here. This is what you see if you are educated in neural physiology. Would it be safe to say that you can only "see" this
45:00 - 45:30 only if you have accepted some theories of neural physiology. There is no way you can come up with this sort of an explanation unless you have a very well elaborated neurophysiological theory. Another way of saying this is that it is not a pure statement of fact. Another way of saying it that this statement is theory-laden, meaning laden by certain neural physiology that you have to accept in order to arrive at that proposition. So it's not a pure statement of fact.
45:30 - 46:00 Now let's take another example. Let's take a favorite telescope if you have one and point it in the direction of the Moon. After a certain time period, you realize that there are mountains on the moon. Is it a pure statement of fact or is this theory-laden? And if this is theory-laden, what sort of theories are presupposed here? Do you understand my quesiton? Is this a pure statement of fact, meaning: am I in a position to say this
46:00 - 46:30 if I put all my theories aside or alternatively: is this based on some tacit theories which I have not yet articulated here? (Student) "By observing the mountains you're placing your trust in the telescople." (Hakob) That's exactly the thing. You have to trust a telescope in order to come up with this proposition, right? You haven't observed this with the naked eye. So you placed your trust in a telescope.
46:30 - 47:00 But in order to trust a telescope, what do you have to assume from the very beginning? A certain optical theory on which the telescope is based, a theory that tells you know that if you use this magnifier and that magnifier and in such combination that is going to give you a trustworthy image. So basically what happens here is our instruments are constructed in accordance with certain theories of optics. We can only accept the results of telescopic observations
47:00 - 47:30 if we trust our optics. This is why this is not a pure statement of fact. This statement is theory-laden. All agree? It is laden by the optical theory that we accept. OK? Now you can see how. But this is still very complex. What about the most basic case like this one. There is an apple in front of me. Is this a pure statement of fact? I'm not using any telescopes here.
47:30 - 48:00 I'm just using naked eye. Is this a pure statement of fact? (Student) "It is based on theories of light and theories about how our eyes work that let's us perceive in the first place?" (Hakob) Very good. So this is the real situation. In reality what happens here is that we do not always trust our sensations. I'm only going to accept
48:00 - 48:30 those perceptions and those sensations which are produced under such-and-such circumstances. Which circumstances? Well, it should be solid elimination: you should be sober, or like your mind is not controlled by any nano-robots. You can easily imagine a situation fifty years from now and it's controlled in such a way that it produces any sort of images. You are not going to trust your perceptions of those.
48:30 - 49:00 We essentially rely on some basic physiology. This may or may not be full-fledged physiological theory. This may be a tacit animal physiology which basically says "trust your senses when there is light." Very basic. You need some basic physiology in order to discriminate between cases when you trust your sense and when you don't trust your senses. How do you know this? Well, you have a physiological theory for that. Again, it might be a very basic theory. It may or may not be articulated; it may or may not the openly stated.
49:00 - 49:30 But nevertheless you have theories or tacit assumptions which allow you to trust your senses. This is what makes this observation theory-laden. We can accept the results of unaided eye observations only if we accept some physiology. Many of you mention other issues: issues with definitions that we need to know what an apple is, etc.. That's all correct. But there is a more fundamental issue:
49:30 - 50:00 any observation is theory-laden. Can we all agree on this? Very good! This bring us to our summary today. This is our summary section. Problem of sensations: we are not in a position to know whether our senses are 100% trustworthy. Problem of induction: no matter how many confirming instances are observed, inductive generalizations are always fallible.
50:00 - 50:30 Finally problem of theory-ladenness: the results of experiments and observations are shaped by all accepted theories. You put these three together and you arrive at very simple conclusion which is the idea of fallibilism: no synthetic proposition can be infallible. Empirical knowledge cannot be absolutely certain. This is the position accepted nowadays, and this is the position that opposes the traditional conception of infallibilism which is the idea that
50:30 - 51:00 empirical science can provide absolute knowledge, that there can be absolutely certain synthetic propositions. A short historical note - a timeline: if you ask Aristotle he would say "yes" that he believed in absolute certainty in empirical science. 2000 years after Aristitole, Immanuel Kant also believed that there can be absolute certainty in physics, chemistry, and biology. He also believed in that. It has been about a hundred years since we come to appreciate that there is no such thing
51:00 - 51:30 as absolute certainty in empirical science. Albert Einstein was just one of many who expressed that idea. It was in that air, the whole generation believed in that. It's safe to say for at least a hundred years we have been fallibilists. We understand that there is no such thing as proof in empirical science. Let's sum it up now. Question at issue is: can there be absolutely certain synthetic proposition? You can say yes and can say no. If you say yes you are an infalliblist,
51:30 - 52:00 meaning that you believe that they can be absolute truth in empirical science and absolutely true synthetic propositions. If you say no, then you are an fallibilist. This is a contemporary position that synthetic propositions cannot be absolutely certain. Now I ask you why do accept this one and not that one? Three reasons: problem of sensations, problem of induction, and problem of theory-ladenness.
52:00 - 52:30 This provides you a summary of today's discussion. One more time, since this is the first time you come across these diagrams: here is the question at issue or the problem. We always start with a problem. In this case it is the question that we're trying to answer. These two here are two opposing viewpoints or conceptions of theses attempting to answer the question. And the finally here we have the reasons why we support or reject a certain position.
52:30 - 53:00 Let's now answer the questions: Can analytic propositions be absolutely certain? Yes. Synthetic? No. What this effectively means is that theories in empirical science such as physics, chemistry, biology, psychology, sociology, political science, economics... they cannot be absolutely certain. The only time when you get absolute certainty is formal science. This is the place when you get perfection.
53:00 - 53:30 There an old joke: when a teacher says "Johnny, what's 2 and 2?" Johnny says "4". Teacher says: "Good!" Johnny says: "What do you mean 'good'. It's perfect!" Perfection is achievable in formal science. Ok, last summary: formal science versus empirical science; Contains only analytic propositions vs. contains synthetic propositions;
53:30 - 54:00 Absolute certainty is achievable and unachievable there; Propositions can be demonstratively proven vs. can only be confirmed by experience but there can be no such thing as proof. Pay attention: there is no such thing as proof in empirical science. For the next time, think about this: if no empirical theory can be absolutely true, what makes us accept one theory and not another theory? We have dismissed the idea of absolute certainty. Dose this mean that any theory is as good as any other?
54:00 - 54:30 Or are we in a position to compare our theories? This is a question we will discuss next time. Thank you very much! Have fun!