The Useless Objective vs. Subjective Distinction
Why is it that people are obsessed with the objective/subjective distinction? You see these terms in all sorts of debates, calling your opponent “subjective” kind of means you won the fight. For example I hear them claim that atheism is subject to radical relativism, or that religious morality is “objective” and that there is no “solid ground to base claims of morality without god”. A lot of arguments revolve around this distinction, I think due to this weird misunderstanding of objectivity and subjectivity. Social construction is conflated with radical relativism, and while I agree there are people who label themselves as relativists who also adhere to social constructivism I do not think that the concept implies a radical relativism of the sort “there are no facts”. No one can actually be a relativist, you simply cannot survive. So we must be clear as to what this means in the context of the objective/subjective distinction and social construction.
Let’s start with the notion of “intelligence”: is it objective or subjective? Well, if objective means “true independent of human existence or opinion”, then in one sense intelligence is not objective in that if humans did not exist, no “thing” could have the property “intelligence” attributed to it (with the exception of animals); but would the animals know they are intelligent? If animals have a level of intelligence but are not self aware, would they know whether intelligence is objective or subjective? Now, let’s put this aside. Let’s call it “objective”, in the sense that “it is there regardless of whether I want it to be or not”. This is the “working” definition I see in conversations that are not rigorous. Have we ever observed intelligence? A cat is objective in the sense that there is an animal here, which I can see and touch and smell and generally perceive; so it objectively exists. But if I say the cat is intelligent; what does that mean? I cannot perceive intelligence directly. I can infer that the concept is present based on behavior. Psychologists use Factor Analysis and structural equation models to infer the presence and level of the latent concept “intelligence”. In other words, there is a theory about “intelligence” and then we try to “see if it is there” based on observational behavioral data. But how are we conceptualizing this “intelligence”? Presumably, we have to define it somehow. This is done by “subjects”; the scientists hypothesizing about it. So is it subjective? Heat is defined as an increase in thermomolecular activity, measurable by observing how many particles are bouncing off one another. What is this definition of intelligence and how are we imposing it upon the world? “Well, scientists are approaching it objectively”. So there are a set of tools, methods, frameworks, and paradigms used when being “objective” about a concept? Let’s assume there is no problem here. Say we agree about all of this; there “is” objective “intelligence” “in” humans. How do I know if someone is “intelligent”? Well, I have to observe them! How am I collecting my data? Is that objective or subjective? After I have collected my data, I need to analyze it and interpret it; this is done by a subject. But how am I doing the interpretation and analysis? Are the methods devised by “subjects”? Where is the “objectivity” here? Let’s suppose a group of scientists agree that a person is intelligent; does consensus mean “objective”? What about correspondence to the world? If something is “hot”, there is by definition an increase in thermomolecular activity. I touch the thing and burn myself. Seems objective to me. If a person is “intelligent”, but they cannot put a spare tire on their car, are they actually “intelligent”? My point with all of these questions is that the objective/subjective distinction makes no difference here. “Intelligence” is only “objective” in the sense that we are trying to measure it “independent of our opinions or feelings or attitudes”. But the community has to agree on what “intelligence” is defined as and how to measure it and what counts as evidence in favor towards the application of the term. It sure sounds like a lot of “subjects” doing the meaning making, and hence not a “transcendent unworldly” “Objective” thing. Surely, few people in the psychological community would say that their work is nothing but objective and if you listen to a person who loves the research, they would also agree that it is “scientifically objective”. What is the difference between Objective and Scientifically Objective; what difference does that qualifier “scientific” do in our understanding of objectivity and subjectivity?
Now let’s turn to “social construction”. People are divided by this term and to what it implies or applies to. Let’s start with an obvious example of a “social fact”. When I go to switch lanes, by law I must turn on my blinker. I turn on the signal indicating to others that I am turning, and they yield to me or cut me off or I wait my turn. It is an “objective fact” that I must do this by law, and “objective fact” that the blinker provides a signal, or communicates an “objective” message, that I am about to turn. Maybe I am doing it for another purpose but let’s ignore that for now. It is also “objective” in a sense that “good” drivers follow the rules of the road such as this one. There is so much to unpack here. Let’s start with the first; it is an objective law that I just turn on my turning signal. Surely this could have been otherwise and is dependent on many other facts and features of the social world. We could live in a society where this does not exist and in other countries and at other times, it did not exist. So is it not objective in that it depends on facts about the social world we have constructed around us? Is it not contingent on where I live and in what era I live? Let’s move to the second objective fact; the meaning given to the blinking signal. Take any blinking signal anywhere: it is objective that it is blinking. But is the meaning we ascribe to it subjective? In one sense no; it is a fact that it is supposed to mean you are turning in that direction. This is objective in the sense that if I die, it will probably still be a fact. It is independent of my opinions. I could change it if I get a ton of people together to overthrow the government and tyranny that is suppressing me. But notice that the “meaning” is also incredibly situational and context bound. A blinking light on a car can mean more than one thing; if a person had their blinker on for a few miles but hasn’t attempted to turn, it probably takes on a new “meaning” that they forgot. When I make this inference, is it subjective? “That they forgot” is the conclusion I have drawn which either is true or false, there is nothing subjective about it. But it is also an inference I am making based on my shared understanding of the world and evidence presented to me at hand. Is inferring necessarily subjective? If that’s the case then all of science would seem to be subjective. Or is it the method I am using to make the inference? I am making a fact claim about my knowledge of the world. “A blinking light factually means that they intend on turning, but they are not turning. What else can explain these conflicts?” , I use my evidence and assumptions about the world to form a conclusion about a fact explaining something. But notice, there are so many layers of “social construction” and “subjectivity” here; the rules of the road are not “objective” like the presence of a cat in front of me, they are entirely dependent on the “social”. It seems like the Objective subjective distinction here is kind of useless. Now what if I say “that person left on their blinker and aren’t turning, therefore they are careless” or say that “they are not a Good driver”. What does any of this mean? We know that if a person puts on a blinker and evades a collision they are in some sense “objectively” a “good” driver. We rate them as somehow “objectively” superior. Insurance companies may even decrease their rates because “objectively” good drivers are less prone to collisions! And “good” drivers are evidenced “objectively” by their lack of collisions and traffic violations! But when I call the other person a “not good” driver, that sure sounds like a “subjective assessment” and that I am giving my “subjective opinion”. Is this not a conflict, or problem, inherent to this subjective objective distinction? The rating system is inherently arbitrary in the sense that if we did not exist, it would not exist, if we rode on horses, our rating standards would not exist. And almost certainly, these ratings were created by some “subject” to aid in assessing the probability of their collision. This sure sounds subjective and a social construction. But at the same time, we can infer the likelihood of different drivers getting into collisions based on data like this! We don’t have to simply state “they are good” or “they suck”, we can take measurements and compare them against past performance! So that sure seems objective, all the while in this realm of social construction and subjectivity. We make this distinctions so haphazardly without understanding how they truly are intertwined. When people say “objective” are they simply masking this complexity and using the word authoritatively? Can something both be a social construction and a fact? Where does subjectivity come into all of this and how does it relate to the all mighty “Objective”?
Moving on to relativism. If a fact is socially dependent, is it not also relative? If an assessment of “good driver” is depends on some standard, is it not also relative to it? Is someone intelligent only relative to the definition agreed upon by those expert psychologists? But I thought intelligence was an objective fact? If something is relative then it must be subjective, right? These are rhetorical questions. Obviously there are elements of objectivity and subjectivity in all events. It is “objective” that a cat is on the couch with me but the experience is entirely subjective. There is no “objective experience of having a cat on you” on the sense of what it makes me feel. Is the “cat” category subjective too? I mean, house cats weren’t always here. And in some sense our society values the little cuties, so does the existence of this category depend on our system of value? Seems subjective and objective are very intertwined.
Bayesian probability is used literally all over the place in science. Some people critique it for being “too subjective” while the Bayesian says that the frequentist paradigm needs to “model its prior information”. Is probability objective or subjective? This is actually a huge topic I do not even want to begin attempting to dive into. Some scientists criticize Bayesian methods, but in time they have crept into scientific communities and have become more accepted. My Economics professors mocked the paradigm while the applied mathematicians were Bayesian purists and made incredible discoveries; making the economics profession look like child’s play. The Bayesian will say that modeling prior information allows us to be more explicit with our assumptions; you can vary these assumptions and see how your conclusions are perturbed under various scenarios. Frequentists can do this somewhat arbitrarily with bootstrapping methods. I don’t want to get into technical details about what paradigm is better but I do want to talk about the usefulness of Bayesian epistemology.
In this notion is the concept of “confidence”. A lot of probability and statistics use this actually. “Confidence intervals” reflect a degree of confidence in the probabilities you measured from a sample of data. I say that I am “confident” in my “ability to drive” meaning that there is a low probability of a collision. Related notions are that of reliability and performance. I don’t “know objectively” if there will be a car crash but I can have varying degrees of confidence based on the information I have. In other words, my assessment of confidence is subjective and depends on the information I have available to me. Is anything objective then if it’s information dependent? Is evidence the key thing making a difference between this arbitrary distinction we love called Objective vs Subjective? When I am doing statistical modeling, this is an “objective” way to make discernments and predictions, but I have to choose the distribution and frame the question, is that not subjective? The probabilities I estimate are typically conditional probabilities meaning they depend on something, are those not objective? Frequentists choose significance levels, what are the objective standards for this? Bayesians weigh incoming evidence relative to current beliefs and revise accordingly. That is what Bayes Rule does, you never get to the “Truth”, you become more and more confident about something as more evidence corroborates or confirms it; but it could always be the case that you are wrong. Assumed in it is the notion that subjectivity is all their is; we are the meaning makers, we are the ones who interpret data and books, and laws and rules. But probably can’t be subjective right? When I role a dice I have an objective 1/6th chance of getting any face value. Surely this is objective! What is the probability of rolling a 3 given the knowledge that the face is odd? 1/3rd!! But what is the probability of a burglar being in my house given the door is unlocked when I normally remember to lock it? This seems like an entirely different question and is not obvious at face value whether the probability we come up with is “objective” or “subjective” but logically the forms are identical; given some knowledge, what is the probability of an unknown? The dice case seems simple because the game is arbitrarily random. It is determined. Your outcomes are already specified, you just have not seen them yet. The other situation is a bit different indeed, and it seems like these different types of uncertainties result in a different qualifier being applied to them: objective or subjective. So, are there subjective elements of science in the sense that “the judgement is subject dependent”; yes. Does that mean subjectivity is valueless? No. Look at the world around us. Note that Bayesianism is about degrees of belief in a proposition, not world views or paradigms. But simpler propositions like hypothesis or aspects of theory. My main point is that subjectivists in the Bayesian view do not think that propositions are not true or false, in other words they are not relativists. They admit that propositions have one truth value, but our degree of belief in whatever that truth value is, depends on information, priors assumptions, background knowledge and is ultimately subjectively assessed. So it is a statement about how we can come to know the world, and aspects about the world, expressed in degrees of confidence and belief which are ultimately formed on the observations we see around us. The dice problem is “objective” in the sense that we have complete information about the process that generates the data, and can effectively estimate the probabilities with certainty. If you know all of the information in the world, then a dice role becomes deterministic. You can predict it’s exact location if you knew everything relevant to the process. “Todays posterior is tomorrows prior” reflects the subjectivist paradigm. My prior assessment of the probabilities will update when I get new information; in other words my subjective assessment of the truth will change as my information base changes. This does not say anything about truth being itself, subjective. This is what it means to be rational according to this epistemology. And if being rational is to be “objective”, then we are updating beliefs in the most rational, logical, and coherent way.
Back to my subjective assessment of the driver who left the blinker on. Given the knowledge of the world I have, I’ve made the assessment that they probably forgot to turn it off. It seems like reasonable, rational, and logical explanation of the observation. Nevertheless someone can dismiss me by saying “you are just subjective, I am Objective and I believe otherwise”. My explanation can be different if I know a bit more information; given that we know people who leave their blinkers on either do by accident or purposefully because they are jokesters, and that I am in a high population of jokers, it’s possible that guy is just messing with us. Or, it is common that people who leave blinkers on do so because they forgot, but I know the guy driving the car and he’s a jokester so therefore maybe he is just messing with us and I can override the alternative explanation. My inference depends on my knowledge. And the degree of reliability of the knowledge. Does that just mean I am being subjective and anyone can dismiss me? I am using evidence, logic, and sound reasoning. I might be wrong but it doesn’t sound unreasonable to believe it, despite it being subjective. I am sure we wouldn’t dismiss all inference on the basis that we are information constrained? Again, the subjective vs objective distinction is kind of useless. And if I judge the driver as “good” or “bad” does that mean it’s entirely subjective and judgement? Supposed my standard is to preserve human life and avoid unnecessary risk, and a person is speeding at 100 mph in a school zone, I’m sure it doesn’t seem subjective to state that he is driving unsafely and is a bad driver. It does seem that I have some form of objective standards, given then likelihood of a crash that is objective on empirical grounds; despite the standards being “socially constructed”.
The point about the blinker is that the simple meaning of a flashing light on your car has objective meaning in a totally socially constructed way; but we would not say that there is no truth. Tomorrow it can be the case that the statement "right blinker on the car indicates a driver intends on turning right" will be false tomorrow if we get some weirdo in office who decides to upend that law. I think what people want to do, is apply the word "objective" to their position, as a form of timeless authority. Think of Boolean logic; it will be true forever that FALSE or FALSE is FALSE, and that TRUE and FALSE is FALSE. See TRUTH TABLES for more. We want this sort of obviousness and authority to apply to our silly beliefs about the physical world. Sorry, it is more difficult then that and empirical work is messy. Get used to it.
Another thing about this approach is that subjective tends to have this connotation that “anything goes”. But In Bayesian epistemology this is not the case, coherence standards are explicit and mathematically proven. Look to the Dutch book arguments for more detail. I think people call other people they disagree with “subjective” as kind of a dig; if you are subjective and incoherent and illogical; and not committed to capital T truth. This is how people use the terms, not in any descriptive sense but in a pejorative way. There is a branch of decision theory utilizing Bayesian methods. This raises an interesting question: can there be “objectively good” decisions? Or can they only be “optimal” with respect to some standard? Are the best “choices” always “subjective”? The distinction is almost always ignored. And useless. There are optimal decisions depending on what you want to maximize.
Terminology to consider:
Possible: pertains to something that could happen, there is nothing inconsistent with logic or metaphysically impossible about the thing. Some people may include possibility as something consistent with physical laws.
Plausible: a subset of possible. There is no statistical data available to make a reasoned conjecture, and not enough information to determine something definitive. Has not met the standards of proof but is not impossible and might even be consistent with common sense or with another background world view. Plausibility and inference to best explanation are related; a defeasible form of reason that is highly assumption laden. Typically these rely on ad hoc hypothesis or other assumptions to make it plausible. This reflects the fact that default explanations are used to fill in the gap when there is missing information. Confirmatory and discomfirmatory evidence should both be considered.
Probable: relating to the frequency or likeliness of the event occurring. Not only is it plausible, but it’s probable: the thing has occurred somewhat regularly to make a judgement and best guess. Something that is probable need not be plausible if the probable outcome is inconsistent with a world view. Nevertheless it is generated with facts and data. The outcome is well defined and the likelihood can be demonstrated.
Proof: there is no room for doubt. The burden of proof has been met with such high precision that it would be foolish to disagree. The evidence points unilaterally to this proposition, or it has been proven apriori simply with the laws of logic and inference. Assumptions are explicit and agreed upon.
Let’s start with the notion of “intelligence”: is it objective or subjective? Well, if objective means “true independent of human existence or opinion”, then in one sense intelligence is not objective in that if humans did not exist, no “thing” could have the property “intelligence” attributed to it (with the exception of animals); but would the animals know they are intelligent? If animals have a level of intelligence but are not self aware, would they know whether intelligence is objective or subjective? Now, let’s put this aside. Let’s call it “objective”, in the sense that “it is there regardless of whether I want it to be or not”. This is the “working” definition I see in conversations that are not rigorous. Have we ever observed intelligence? A cat is objective in the sense that there is an animal here, which I can see and touch and smell and generally perceive; so it objectively exists. But if I say the cat is intelligent; what does that mean? I cannot perceive intelligence directly. I can infer that the concept is present based on behavior. Psychologists use Factor Analysis and structural equation models to infer the presence and level of the latent concept “intelligence”. In other words, there is a theory about “intelligence” and then we try to “see if it is there” based on observational behavioral data. But how are we conceptualizing this “intelligence”? Presumably, we have to define it somehow. This is done by “subjects”; the scientists hypothesizing about it. So is it subjective? Heat is defined as an increase in thermomolecular activity, measurable by observing how many particles are bouncing off one another. What is this definition of intelligence and how are we imposing it upon the world? “Well, scientists are approaching it objectively”. So there are a set of tools, methods, frameworks, and paradigms used when being “objective” about a concept? Let’s assume there is no problem here. Say we agree about all of this; there “is” objective “intelligence” “in” humans. How do I know if someone is “intelligent”? Well, I have to observe them! How am I collecting my data? Is that objective or subjective? After I have collected my data, I need to analyze it and interpret it; this is done by a subject. But how am I doing the interpretation and analysis? Are the methods devised by “subjects”? Where is the “objectivity” here? Let’s suppose a group of scientists agree that a person is intelligent; does consensus mean “objective”? What about correspondence to the world? If something is “hot”, there is by definition an increase in thermomolecular activity. I touch the thing and burn myself. Seems objective to me. If a person is “intelligent”, but they cannot put a spare tire on their car, are they actually “intelligent”? My point with all of these questions is that the objective/subjective distinction makes no difference here. “Intelligence” is only “objective” in the sense that we are trying to measure it “independent of our opinions or feelings or attitudes”. But the community has to agree on what “intelligence” is defined as and how to measure it and what counts as evidence in favor towards the application of the term. It sure sounds like a lot of “subjects” doing the meaning making, and hence not a “transcendent unworldly” “Objective” thing. Surely, few people in the psychological community would say that their work is nothing but objective and if you listen to a person who loves the research, they would also agree that it is “scientifically objective”. What is the difference between Objective and Scientifically Objective; what difference does that qualifier “scientific” do in our understanding of objectivity and subjectivity?
Now let’s turn to “social construction”. People are divided by this term and to what it implies or applies to. Let’s start with an obvious example of a “social fact”. When I go to switch lanes, by law I must turn on my blinker. I turn on the signal indicating to others that I am turning, and they yield to me or cut me off or I wait my turn. It is an “objective fact” that I must do this by law, and “objective fact” that the blinker provides a signal, or communicates an “objective” message, that I am about to turn. Maybe I am doing it for another purpose but let’s ignore that for now. It is also “objective” in a sense that “good” drivers follow the rules of the road such as this one. There is so much to unpack here. Let’s start with the first; it is an objective law that I just turn on my turning signal. Surely this could have been otherwise and is dependent on many other facts and features of the social world. We could live in a society where this does not exist and in other countries and at other times, it did not exist. So is it not objective in that it depends on facts about the social world we have constructed around us? Is it not contingent on where I live and in what era I live? Let’s move to the second objective fact; the meaning given to the blinking signal. Take any blinking signal anywhere: it is objective that it is blinking. But is the meaning we ascribe to it subjective? In one sense no; it is a fact that it is supposed to mean you are turning in that direction. This is objective in the sense that if I die, it will probably still be a fact. It is independent of my opinions. I could change it if I get a ton of people together to overthrow the government and tyranny that is suppressing me. But notice that the “meaning” is also incredibly situational and context bound. A blinking light on a car can mean more than one thing; if a person had their blinker on for a few miles but hasn’t attempted to turn, it probably takes on a new “meaning” that they forgot. When I make this inference, is it subjective? “That they forgot” is the conclusion I have drawn which either is true or false, there is nothing subjective about it. But it is also an inference I am making based on my shared understanding of the world and evidence presented to me at hand. Is inferring necessarily subjective? If that’s the case then all of science would seem to be subjective. Or is it the method I am using to make the inference? I am making a fact claim about my knowledge of the world. “A blinking light factually means that they intend on turning, but they are not turning. What else can explain these conflicts?” , I use my evidence and assumptions about the world to form a conclusion about a fact explaining something. But notice, there are so many layers of “social construction” and “subjectivity” here; the rules of the road are not “objective” like the presence of a cat in front of me, they are entirely dependent on the “social”. It seems like the Objective subjective distinction here is kind of useless. Now what if I say “that person left on their blinker and aren’t turning, therefore they are careless” or say that “they are not a Good driver”. What does any of this mean? We know that if a person puts on a blinker and evades a collision they are in some sense “objectively” a “good” driver. We rate them as somehow “objectively” superior. Insurance companies may even decrease their rates because “objectively” good drivers are less prone to collisions! And “good” drivers are evidenced “objectively” by their lack of collisions and traffic violations! But when I call the other person a “not good” driver, that sure sounds like a “subjective assessment” and that I am giving my “subjective opinion”. Is this not a conflict, or problem, inherent to this subjective objective distinction? The rating system is inherently arbitrary in the sense that if we did not exist, it would not exist, if we rode on horses, our rating standards would not exist. And almost certainly, these ratings were created by some “subject” to aid in assessing the probability of their collision. This sure sounds subjective and a social construction. But at the same time, we can infer the likelihood of different drivers getting into collisions based on data like this! We don’t have to simply state “they are good” or “they suck”, we can take measurements and compare them against past performance! So that sure seems objective, all the while in this realm of social construction and subjectivity. We make this distinctions so haphazardly without understanding how they truly are intertwined. When people say “objective” are they simply masking this complexity and using the word authoritatively? Can something both be a social construction and a fact? Where does subjectivity come into all of this and how does it relate to the all mighty “Objective”?
Moving on to relativism. If a fact is socially dependent, is it not also relative? If an assessment of “good driver” is depends on some standard, is it not also relative to it? Is someone intelligent only relative to the definition agreed upon by those expert psychologists? But I thought intelligence was an objective fact? If something is relative then it must be subjective, right? These are rhetorical questions. Obviously there are elements of objectivity and subjectivity in all events. It is “objective” that a cat is on the couch with me but the experience is entirely subjective. There is no “objective experience of having a cat on you” on the sense of what it makes me feel. Is the “cat” category subjective too? I mean, house cats weren’t always here. And in some sense our society values the little cuties, so does the existence of this category depend on our system of value? Seems subjective and objective are very intertwined.
Bayesian probability is used literally all over the place in science. Some people critique it for being “too subjective” while the Bayesian says that the frequentist paradigm needs to “model its prior information”. Is probability objective or subjective? This is actually a huge topic I do not even want to begin attempting to dive into. Some scientists criticize Bayesian methods, but in time they have crept into scientific communities and have become more accepted. My Economics professors mocked the paradigm while the applied mathematicians were Bayesian purists and made incredible discoveries; making the economics profession look like child’s play. The Bayesian will say that modeling prior information allows us to be more explicit with our assumptions; you can vary these assumptions and see how your conclusions are perturbed under various scenarios. Frequentists can do this somewhat arbitrarily with bootstrapping methods. I don’t want to get into technical details about what paradigm is better but I do want to talk about the usefulness of Bayesian epistemology.
In this notion is the concept of “confidence”. A lot of probability and statistics use this actually. “Confidence intervals” reflect a degree of confidence in the probabilities you measured from a sample of data. I say that I am “confident” in my “ability to drive” meaning that there is a low probability of a collision. Related notions are that of reliability and performance. I don’t “know objectively” if there will be a car crash but I can have varying degrees of confidence based on the information I have. In other words, my assessment of confidence is subjective and depends on the information I have available to me. Is anything objective then if it’s information dependent? Is evidence the key thing making a difference between this arbitrary distinction we love called Objective vs Subjective? When I am doing statistical modeling, this is an “objective” way to make discernments and predictions, but I have to choose the distribution and frame the question, is that not subjective? The probabilities I estimate are typically conditional probabilities meaning they depend on something, are those not objective? Frequentists choose significance levels, what are the objective standards for this? Bayesians weigh incoming evidence relative to current beliefs and revise accordingly. That is what Bayes Rule does, you never get to the “Truth”, you become more and more confident about something as more evidence corroborates or confirms it; but it could always be the case that you are wrong. Assumed in it is the notion that subjectivity is all their is; we are the meaning makers, we are the ones who interpret data and books, and laws and rules. But probably can’t be subjective right? When I role a dice I have an objective 1/6th chance of getting any face value. Surely this is objective! What is the probability of rolling a 3 given the knowledge that the face is odd? 1/3rd!! But what is the probability of a burglar being in my house given the door is unlocked when I normally remember to lock it? This seems like an entirely different question and is not obvious at face value whether the probability we come up with is “objective” or “subjective” but logically the forms are identical; given some knowledge, what is the probability of an unknown? The dice case seems simple because the game is arbitrarily random. It is determined. Your outcomes are already specified, you just have not seen them yet. The other situation is a bit different indeed, and it seems like these different types of uncertainties result in a different qualifier being applied to them: objective or subjective. So, are there subjective elements of science in the sense that “the judgement is subject dependent”; yes. Does that mean subjectivity is valueless? No. Look at the world around us. Note that Bayesianism is about degrees of belief in a proposition, not world views or paradigms. But simpler propositions like hypothesis or aspects of theory. My main point is that subjectivists in the Bayesian view do not think that propositions are not true or false, in other words they are not relativists. They admit that propositions have one truth value, but our degree of belief in whatever that truth value is, depends on information, priors assumptions, background knowledge and is ultimately subjectively assessed. So it is a statement about how we can come to know the world, and aspects about the world, expressed in degrees of confidence and belief which are ultimately formed on the observations we see around us. The dice problem is “objective” in the sense that we have complete information about the process that generates the data, and can effectively estimate the probabilities with certainty. If you know all of the information in the world, then a dice role becomes deterministic. You can predict it’s exact location if you knew everything relevant to the process. “Todays posterior is tomorrows prior” reflects the subjectivist paradigm. My prior assessment of the probabilities will update when I get new information; in other words my subjective assessment of the truth will change as my information base changes. This does not say anything about truth being itself, subjective. This is what it means to be rational according to this epistemology. And if being rational is to be “objective”, then we are updating beliefs in the most rational, logical, and coherent way.
Back to my subjective assessment of the driver who left the blinker on. Given the knowledge of the world I have, I’ve made the assessment that they probably forgot to turn it off. It seems like reasonable, rational, and logical explanation of the observation. Nevertheless someone can dismiss me by saying “you are just subjective, I am Objective and I believe otherwise”. My explanation can be different if I know a bit more information; given that we know people who leave their blinkers on either do by accident or purposefully because they are jokesters, and that I am in a high population of jokers, it’s possible that guy is just messing with us. Or, it is common that people who leave blinkers on do so because they forgot, but I know the guy driving the car and he’s a jokester so therefore maybe he is just messing with us and I can override the alternative explanation. My inference depends on my knowledge. And the degree of reliability of the knowledge. Does that just mean I am being subjective and anyone can dismiss me? I am using evidence, logic, and sound reasoning. I might be wrong but it doesn’t sound unreasonable to believe it, despite it being subjective. I am sure we wouldn’t dismiss all inference on the basis that we are information constrained? Again, the subjective vs objective distinction is kind of useless. And if I judge the driver as “good” or “bad” does that mean it’s entirely subjective and judgement? Supposed my standard is to preserve human life and avoid unnecessary risk, and a person is speeding at 100 mph in a school zone, I’m sure it doesn’t seem subjective to state that he is driving unsafely and is a bad driver. It does seem that I have some form of objective standards, given then likelihood of a crash that is objective on empirical grounds; despite the standards being “socially constructed”.
The point about the blinker is that the simple meaning of a flashing light on your car has objective meaning in a totally socially constructed way; but we would not say that there is no truth. Tomorrow it can be the case that the statement "right blinker on the car indicates a driver intends on turning right" will be false tomorrow if we get some weirdo in office who decides to upend that law. I think what people want to do, is apply the word "objective" to their position, as a form of timeless authority. Think of Boolean logic; it will be true forever that FALSE or FALSE is FALSE, and that TRUE and FALSE is FALSE. See TRUTH TABLES for more. We want this sort of obviousness and authority to apply to our silly beliefs about the physical world. Sorry, it is more difficult then that and empirical work is messy. Get used to it.
Another thing about this approach is that subjective tends to have this connotation that “anything goes”. But In Bayesian epistemology this is not the case, coherence standards are explicit and mathematically proven. Look to the Dutch book arguments for more detail. I think people call other people they disagree with “subjective” as kind of a dig; if you are subjective and incoherent and illogical; and not committed to capital T truth. This is how people use the terms, not in any descriptive sense but in a pejorative way. There is a branch of decision theory utilizing Bayesian methods. This raises an interesting question: can there be “objectively good” decisions? Or can they only be “optimal” with respect to some standard? Are the best “choices” always “subjective”? The distinction is almost always ignored. And useless. There are optimal decisions depending on what you want to maximize.
My point is that we should be critical of this distinction and weary of its practical application. John Searle points out that our common conception of the distinction is fundamentally flawed. He points it out here, in the "Two Varieties of the Subjective-Objective Distinction:
Epistemology is about "how we know" and Ontology is about "What there is". Experiences are Ontologically subjective in the sense that they exist within a subject; they are subject dependent. This is different than Ontologically objective entities such as physical structures that are independent of human existence. Epistemological objectivity is about statements that can be verified in principle, while subjectivity falls within the domain of a subjects personal assessment of something. Why is this important? Because people consistently equivocate the two senses, not making the necessary distinction between these senses during a discussion. Based on my argument earlier about "good drivers", I would further break down the subjective epistemological category into a section containing instances where we (subjectively) define standards about "goodness" associated with an action or behavior, which we can then measure observations against. It is Epistemically subjective that we should avoid catastrophe. However, once we start with this axiom, we can begin to state epistemically objective propositions about whether certain actions, strategies, or behaviors are conducive to the goal. I believe this is in fact how all subjective epistemological discussions occur. Of course, there is nothing ontologically objective about what axiom we choose. These are typically presumptions that are a function of culture and socialization. My question; is there an axiomatic starting point that all rational beings can accept, so we can reason normatively about, and orient our goals towards, a common framework for evaluating epistemically subjective statements, so that they become ontologically subjective and subject to more rigorous analysis? We do this with Stop Signs and "Good Drivers". We even do it with Money; it is nothing more than a socially agreed upon convention that persists across generations. You will definitely not hear a person reject a an offer of 100$ on grounds that it is a social construction. Nor will you rarely see a person blow through a stop sign at 100MPH. There has to be some fact of the matter, given a context and rules of engagement. The question of whether the rules are sub-optimal relative to a collection of goals doesn't seem entirely subjective to me. Can certain goals that seem to persist across space and time, available in even seemingly divergent civilizations, be considered ontologically subjective?
My pains have a subjective mode of existence in that they only exist as experienced by me, the subject. But mountains and molecules have an objective mode of existence because they exist whether or not they are experienced by any subject. It can be an epistemically objective matter of fact that I have a pain even though the mode of existence of the pain is ontologically subjective. (Searle. Philosophy in a New Century. New York: Cambridge, 2008.167)
Impartiality and neutrality are obviously two closely related concepts; especially under the law. Can a mediator effectively stand unbiased when resolving a dispute between two parties who have diverging interests? Does a random selection of a jury guarantee an impartial assessment of "the facts"? It seems like what someone deems as relevant information in any situation is determined in-part by a range of factors that call into question the objectivity of the selection process. Search neutrality is a big topic related to this. How do we prioritize search results on Google free from privileging certain sources over others, while minimizing the amount of user-time spent searching for the relevant material? It seems like whatever choice we make, we are arbitrarily privileging certain information over others; even if we alphabetically order the results this gives preference to some and not others. This is not a trivial topic, just food for thought. See more about Algorithmic Bias elsewhere; I bring it up because it just seems relevant to the current topic because many people reaffirm what they think to be ontologically objective truths on the basis of an incredibly partial search result or newsfeed. News should be boring, but instead its as sensational and thrilling as gambling. This notion of "Objectivity" has been discussed in the context of moral reasoning. Adam Smith developed the notion of an "Impartial Spectator" who can be free of sympathetic sentiments and only concerned with "The Facts"; is it actually possible to become the impartial spectator as envisioned by Smith and other liberal thinkers? The Veil of Ignorance is another concept that invokes the notion of impartiality. Is it possible to "Stand outside" of our socio-economic condition in such a way to be free of bias? For some, most of these questions reduces to a form of radical subjectivity; everything is a function of power structures. I think this is unfortunate. For me the problem is akin to that of local optimization and error correction; given the known mechanisms by with partiality and bias arise, try and correct them. Because we don't know all of the ways we can fool ourselves, have an error correction feedback loop that filters out false positives, while continuing an open ended dialogue to identify the global optima. Finding truth requires being honest with ourselves about our fundamental self-affirming tendencies; its okay to be wrong and uncertain.
I think its also important to note that radical relativism is not a uniquely "politically left" phenomenon. Rather, you can think of it as inherently "illiberal"; you can be illiberal in all directions of the political spectrum. Consider someone like Aleksander Dugin, his political theory seems to be the guiding hand for the Russian government. He firmly advocates for a "post-modern" analysis of geopolitics; claiming that there "is a Russian truth". He is on the right of the political aisle on absolutely every issue and yet he is a radical relativist; he is a far-right relativist who dismisses all of Western Civilization (illiberal; like Mao and Hitler). If you value coherence and consistency you should probably be equally dismissive of any form of radical relativism regardless of the specific variants political orientation. In any case, it seems intellectually pathetic when someone resorts to "its my truth" or falls back on standpoint epistemology when engaging in matters of fact. There are all sorts of people who pull this bullshit, it does not matter their political orientation. Everyone should be weary of it and apply the intellectual standards (see below) to all propositions (including your own):
Anyway I just wanted to explore these concepts because it seems like people just use them to take the high ground in an argument. This is not to say that knowledge is not attainable or anything like that, it’s just a lot harder when you actually sit down and think about it. Even our interpretation of objective measurement requires some sort of subjectivity; consider false positives of binary tests and the data collection processes. This is also not to say that examples of subjectivity and objectivity do not exist. It’s pretty subjective for someone to just assert a belief without reason and it’s pretty objective for me to see an axiomatic mathematical proof of the binomial theorem. I would just suggest keeping the terms in these clear domains and stop throwing them around outside of these contexts. The point being that I just see them as argumentative tactics. There is a difference between what is stated and what is meant. When we use these words we are communicating something else under the pretense of scholarship. They appear to be carefully selected phrases with a specific Gricean implicature. They serve pragmatic purposes in a specific speech act; not something we can use to advance a discussion towards truth.
Resource:
Resource:
2. Implicature
6. Normativity
Terminology to consider:
Possible: pertains to something that could happen, there is nothing inconsistent with logic or metaphysically impossible about the thing. Some people may include possibility as something consistent with physical laws.
Plausible: a subset of possible. There is no statistical data available to make a reasoned conjecture, and not enough information to determine something definitive. Has not met the standards of proof but is not impossible and might even be consistent with common sense or with another background world view. Plausibility and inference to best explanation are related; a defeasible form of reason that is highly assumption laden. Typically these rely on ad hoc hypothesis or other assumptions to make it plausible. This reflects the fact that default explanations are used to fill in the gap when there is missing information. Confirmatory and discomfirmatory evidence should both be considered.
Probable: relating to the frequency or likeliness of the event occurring. Not only is it plausible, but it’s probable: the thing has occurred somewhat regularly to make a judgement and best guess. Something that is probable need not be plausible if the probable outcome is inconsistent with a world view. Nevertheless it is generated with facts and data. The outcome is well defined and the likelihood can be demonstrated.
Proof: there is no room for doubt. The burden of proof has been met with such high precision that it would be foolish to disagree. The evidence points unilaterally to this proposition, or it has been proven apriori simply with the laws of logic and inference. Assumptions are explicit and agreed upon.
Comments
Post a Comment