How is it possible that experts in a certain field can disagree despite given access to the same facts?

For example in science, why is there disagreement when given access to the exact same information? Shouldn't the scientific method correct the information so that it is at least similar? Thank you for your time

Edit: This is kind of a secondary question (very basic philosophy knowledge layman), but why are there different interpretations of different scientific facts? What causes these differences within people's interpretations and decision to accept certain theories?

Examples: Scientists disagreeing over string theory, biologists disagreeing over a scientific paper, or scientists disagreeing over global warming. Art critics disagreeing over the quality of a painting, classical musicians disagreeing over an interpretation of a song.

  • 2
    Because "complex" facts (like those considered by science) are hardly "intelelgible" without interpretations. Commented Sep 10, 2016 at 18:25
  • 1
    because all "facts" are theoery-laden. info != fact
    – user20153
    Commented Sep 10, 2016 at 20:25
  • well, if I pick up a stone and drop it; we can both agree that it falls, we might disagree why it falls; this happens all the time... Commented Sep 10, 2016 at 21:09
  • 3
    Hi. Could you add to the question an example of such a disagreement? Commented Sep 10, 2016 at 21:58
  • Hi, welcome to philosophy SE. The reason is similar to why we can draw multiple curves through the same collection of dots. Science is like a curve connecting the dots of facts, but experts may disagree as to the "best" ways to connect them. Those rely on informal heuristics and are not prescribed by the scientific, or any other general method.
    – Conifold
    Commented Sep 10, 2016 at 22:02

13 Answers 13


Do not think of the scientific method as a process of deduction.

Nor should it be thought of as only used on typical science topics.

Take any complex system and you can apply the scientific method. For this example we can use an automobile.

Let's say an automobile has a problem - it is getting poor gas mileage and has low power. Many problems can cause this symptom, and you can develop theories on what the problem(s) are. For instance, a clogged exhaust, exhaust leak, leaky air intake, bad air intake sensor, poor spark plugs, etc. are all possible causes. Once you add data, it may begin to narrow down these theories. Different theories will explain data in different ways (for this application, intake and exhaust theories will have good discussions on fuel trim data).

The same is applicable to science. Geocentric models could explain orbits for a long time - but heliocentric models made the explanation much simpler. Different models of the atom existed, and some models could predict some real world behaviour - until new data from a gold foil experiment left a single theory as the most plausible.

You will see these disagreements occur mainly over incomplete data. A single fact of data can be explained by different theories in different ways. Its not a deductive process. To learn more, I recommend researching the historic development of scientific theories.


As stated, most of your questions don't strike me as being philosophical ones. How is it possible that experts, say scientists, disagree? Well, presumably this is explained by psychological facts about the scientists in question--for example, although they're experts, they're still human beings, so fallible and influenced by irrational biases and prejudices--and facts about the complexity of the domain in question--physics is hard; it's not always straightforward what theories the experimental data support, for instance.

There are some philosophical questions in the neighborhood of the questions you've asked, however. Could two disagreeing experts both be rational? Epistemologists have been interested in this question for quite awhile. If you think that the total evidence uniquely determines which beliefs are rational, you'd probably be inclined to answer the question in the negative. But if you think that evidence can be indeterminate in some way, then you might answer in the affirmative.

ADDENDUM: Let me sketch for you a model of rational scientific disagreement that is actually prevalent in the philosophy of science, epistemology, and decision theory literature, since most of the current responses seem rather far off the mark to me. The model I'll sketch is sometimes called Bayesianism.

According to Bayesianism, a rational inquirer begins inquiry with a probability distribution, called a prior, over the hypotheses of interest to her. The prior represents the inquirer's degrees of belief in the hypotheses before collecting any evidence. Orthodox Bayesianism (DeFinetti, Savage, etc.) is extremely subjective in the sense that rationality only requires that these prior distributions actually be probability measures (as opposed to arbitrary set functions). As inquirers collect evidence, they update their priors by a process called conditionalization. Roughly, their updated probability distributions are equal to their prior conditional distributions given the evidence. This is a highly simplified exposition, but will suffice to answer your question.

Now, suppose this model is right, and let us ask again how two rational scientists might disagree on this model. The answer is simple. If Scientist 1 and Scientist 2 have different priors, then, even if they update by conditionalizing on exactly the same evidence, in general their updated probability distributions will be different. On this model, rationality demands only that priors be probabilistically coherent and that they be updated by conditionalization. Since for a given (non-trivial) space of hypotheses there are infinitely many coherent probability measures on that space, it's easy to find examples of rational disagreement.

  • By what conditions, specifically, is this question rendered non-philosophical? Commented Sep 12, 2016 at 20:40

Why do scientists disagree over the same given fact?

  1. Scientists might also disagree over whether to count what they see as the "same" one. That is, they might argue whether their different observations equal one same fact.
    • It seems you have in mind a picture like this: a fact is given as an ingredient. There is a recipe to cook the fact. If we want the best (the truth about nature) out of it, why don't we just follow the best recipe available in the same way?
    • But notice: there are even many ways to characterize a fact. For example, it might be just one of many ways to express phenomena that you "see" a "stone" "falling". Is the stone "falling"? What does that even mean?
  2. Even if they set aside the problem of defining the "same fact", scientists may have different "agendas", i.e., what they want to achieve with their theories.
  3. We can assume: Scientists want their theory to (a) predict future events well, (b) show how things become so, and (c) give us an intelligible explanation for why they are so. And scientists often disagree which theory is the best in terms of (a), (b), and (c).
    • That is, scientists have to choose a theory to explain the given fact. scientists disagree which theory best accommodates the given fact. They see many ways to fit the fact into different theories.
  4. Who knows? Maybe scientists disagree just because they don't like each other. This may sound ridiculous but it happens very often. (This is one of many explanations -- which are not mutually exclusive -- of "why" scientists often disagree.)
  • 2
    Welcome to philosophy and thanks for the answer! Is there any chance you might be able to unpack this a little further? (Why is this a persuasive answer to the question? What research could confirm it?)
    – Joseph Weissman
    Commented Sep 14, 2016 at 2:27
  • 1
    @JosephWeissman, thank you for your kind comment. I am not sure if I can provide a list of further references, because my answer is full of questions, which are too broad, vague, and ambiguous. (Maybe that's why you felt I needed to explain why this is a persuasive answer.) I meant my answers to be vague enough, though: the original question itself, at the time I read, seemed to be touching on many different topics -- not just philosophy of science and epistemology but also sociology and aesthetics, etc. But yeah, I will soon come up with a more concrete and informative answer, though! Commented Sep 15, 2016 at 15:29

The answers above cover much ground, but I hope I can add some names to the points that have been made for your further reading.

Scientific investigation nowadays mostly follows the methods of Karl Popper who, in summary, suggests that scientific theories cannot be deduced from the facts, they are arrived at abstractly and then tested against the facts. Those that can be falsified are rejected, those that remain we believe until they too are falsified. This alone leads directly to an answer in that the fact does not produce the theory, the theory comes first, the fact then either falsifies it or not. Given that falsification is rarely complete (i.e. our data is often incomplete), the views of scientists about new or emerging theories depend more often on what they have chosen as their working hypothesis, than the facts against which they are still testing them.

Further insight can be gained from the work of Thomas Kuhn who introduced an element of the neuroscientific theory of cognitive dissonance to the scientific process. By this theory, people experience pain when faced with information which substantially contradicts what they currently believe. Results which should falsify a theory currently held to be true (the current paradigm), are often interpreted as errors or even deliberately misread, in order to avoid the pain of cognitive dissonance. Thus differences in the theories of scientists can be explained by their different responses to the pain of cognitive dissonance.

Early pragmatist philosophers like Charles Peirce had previously postulated a version of Popper's falsification, considering (to varying degrees) the "truth" to be more-or-less what works. Thus attempts to falsify theories will also depend to a certain extent on a scientist's view as to whether the theory is "working" or not.

Finally, a significant problem nowadays (much less relevant in the past) is that scientists gain reputation by publication and the editors of major science journals are only interested in papers which are new or exciting. This leads to a drive to investigate the new rather than confirm the existing. This feature of modern science has been widely explored, an example would be the inflammatorily titled (but nonetheless interesting) paper "Why Most Published Research Findings Are False" by John Ioannidis" (I add this paper with trepidation, please don't interpret as "all science is wrong" it definitely does not say that).

In summary, scientists attempt to use Popper's falsification approach, which leads to a few theories all being around at the same time, having not been fully falsified yet. Cognitive dissonance (as well as other human weakness) leads to some theories hanging on for longer than they should while others push the boundaries, and the modern publication regime encourages new or exciting theories over re-affirming old ones, some are more tempted by this than others.


There is, in fact, a lot of work in social epistemology on precisely this question.

In short, if the agents are epistemic peers -- i.e. equally good at reasoning, with the same evidence etc -- then there are two views people take on learning of their disagreement. The Equal Weight View has it that the agents ought to adopt some sort of "compromise" position in between their original positions. This seems to not reflect the fact of widespread intransigent disagreements. The alternative is the Right Reasons View which says that each should stick to their guns and not change their mind. This seems to accommodate the facts of there actually being disagreements between peers, but, in my view, seems unmotivated.


I think what the others are getting at is this: in science there is no disagreement over the facts, only the theory. If we take coin-flipping to be a form of science, and we flip coins repeatedly with a group of scientists present, everyone will agree about a few basic things like how many heads they observed or what class of material the coin is made of. The problem of disagreement comes when they learn more about the world from this experiment.

Some of our scientists might develop a theory: that if this coin is fair, it should fall on heads as often as tails in the limiting case. Some others might notice that the coin was flipped 50 times and argue that the coin is likely biased, as it fell on heads 35 times instead of 25. Some may say that this is reasonably unlikely, and thus proves the coin's bias, while others might say we need to flip more to come to a conclusion.

Others might attempt to model the coin's behavior to understand the laws that govern its motion. These models will likely have varying degrees of accuracy and depend on other variables like the exact chemical composition of the coin - leading the chemists of the group to experiment with samples of the coin to understand its nature.

Others still will wonder what in the world I'm flipping a coin for and will seek to analyze my role, or even their own as a member of this social process, theorizing about our underlying motivations for performing these acts when there are other groups who care nothing for the behavior of coins.

I could really go on and on here but hopefully you've got the point - theories are the subject of disagreement, not the base level facts. Theories are falsifiable but not provable (by definition) - there can be no theory in science immune to the possibility of future rejection. Thus, the disagreements are often along the lines of whether a theory has been falsified, or sometimes whether the theory is falsifiable or not.

  • 1
    This seems quite wrong to me. First, what exactly are the "base level facts" that you have in mind? Do you have in mind outcomes of experiments? If so, scientists can certainly disagree about these. Consider, for instance, an experiment with CERN's Large Hadron Collider. In order to even say what the data are, many modeling assumptions need to be made. Scientists who prefer somewhat different statistical models than the ones actually used may have disagreements about the outcomes of experiments. The problem with your response is that coin flipping is a far too simplistic example.
    – user22273
    Commented Sep 12, 2016 at 23:13
  • 1
    "there can be no theory in science immune to the possibility of future rejection." I should add that you don't take this Quinean point far enough. Quine was a holist about, not just about "theories", but about all beliefs. Even the "base level facts" you talk about are not incorrigible. (Some holists may argue that logical truths are incorrigible, but certainly not empirical statements.)
    – user22273
    Commented Sep 12, 2016 at 23:14
  • Yes, it does leave the concept of measurement error out of the equation, and it is an overly simple example that needn't invoke measurement or the role of statistics in science. From a theoretical standpoint, you're absolutely right that the "base level facts" are not out of the question. However, it's also true that without good reason to dismiss these "base facts", you will be laughed out of your laboratory. It's not simply enough to say "There's no way this data could be correct!" and move on with your life, similar to how philosophers must justify their arguments to be taken seriously. Commented Sep 13, 2016 at 4:01
  • I agree with what you just wrote, but I don't see how it answers my points. You wrote, "there is no disagreement over the [base level?] facts, only the theory." I deny this. You also wrote, " theories are the subject of disagreement, not the base level facts," which I also deny. I don't think I ever claimed that beliefs in base facts (or anything, for that matter) can or should be dismissed without good reason. I wouldn't subscribe to that view.
    – user22273
    Commented Sep 13, 2016 at 4:08
  • It answers your points in that I'm saying, yes, there is more nuance to this, and yes, I saved my fingers some work by not addressing every issue related to philosophy of science in writing it. I also used such an example because it caters to the original question (and its asker's level of depth)- "How can scientists disagree given the same base facts?" Maybe what I'm getting at is this: in order to throw out the facts, you must also do some science that shows the measurements to be wrong - in effect, just adding another layer of facts to the equation or generating new data points. Commented Sep 13, 2016 at 4:26

They can disagree on what weight they put on the facts or even if they're relevant at all to what is shown as the findings from those facts.

For example, maybe some scientist has data that shows that rabbits are going to take over the world in 4 years and we're all doomed. He has accurate statistics about growth of rabbits in Siberia. He also has a company selling a solution to the rabbit problem. The same data is given to other scientists who use their life experience to say that it's bullcrap due to there going to be factors that limit the growth of that group of rabbits - the facts of growth of that rabbit group are still the same.

Also you don't need to be a scientist to take advantage of the chewbecca defense, where you just present some facts(that anyone can understand) and then provide a conclusion that has nothing to do with the presented facts, leading to guaranteed disagreement - but a layman observing the disagreement might not understand why, as the facts presented are the same, it's just that the layman can understand the presented facts but doesn't really understand how they are not connected at all to the disagreement.


Turn this around. The deeper mystery, how is it possible that anyone agrees? How can we set up a process for creating agreement by the most people, over the widest range of things, through a process of reasoned argument? Simply put, that quest gives you science.

Sure as people have said personal issues get in the way. Look at Eddington using his prestige against Chandrasekhar. But these are failurures in the process. While of course, real meaningful disagreements occur when the method is used correctly.

There are different kinds of facts, in different areas of science. A new particle is pretty unambiguous. But global warming and whether it is human-generated, that has been a process of building degrees of consensus. We develop processes for agreeing on and verifying facts. Science never truly proves anything, it only evaluates degrees of uncertainty.

The conventional view is classical empiricism and the classical observationalist-inductivist account based on it. Anyone can do the experiments, or examine how the data was gathered, and the best theory will account for the most facts with the greatest simplicity. Anything that doesn't agree with the facts, gets falsified. This often hits problems as a model. For instance 'simplicity' is not a neutrally agreeable on criteria, we look for elegance, something aesthetic.

Popper put the onus on the scientific community, to be able to verify the facts, which hypothesees about them are valid, and to choose between them. So the disagreements would come down to disagreements on how evidence was obtained, what it is evidence for, and how strongly an interpretatiin could be drawn distinctly from other possible interpretations; all within the context of the scientific community, tryingbto make pursuasive cases, like we do on Philosophy SE.

Kuhn argued that observation itself is theory laden, and developed a model of paradigm change based on observations of development of actual scientific practice. People with different beliefs, operating in different paradigms, might report the same observations in very different ways.

People who hang their whole careers on an area of research, certainly get very invested in that. String theory and loop quantum gravity can't both be right. It will take the convergence of detailed enough theory to meet observations of sufficiently high energies and small scales to decide - eg black holes merging. People's life's work matter to them, but this is corrected for by the fact prestige is equally or more generated by knocking down the ideas of others, as building new ones up.

Biology is a lot more prone to methadological disagreements. Behaviourists looking at drug to food preference of rats in skinner boxes and expecting pavlovian responses anyway, came to very different conclusions about addiction than later researches who compared behaviour in skinner boxes to a 'rat park', a more normal healthy social situation for rats, instead of solitary confinement. It has taken a really long time for the insights from this later work, maybe 40 years, for this to start to affect government policies. This has been even more politicised than global warming. Social context, and funding, matter.


In every case, it is because they either are holding on to a religion or they are not experts. It could be the standard model, it could be heliocentrism (which in fact has never been observed), it could be their belief that Jesus is GOD, affecting them.

Whatever the case, the fact that you haven't seen this means that YOU also are a member of the religions in questions. Be a true scientist and scrutinize every model. Did they measure the mass of the planets. Nope. How deep of well would be required to measure 93M miles of distance to the sun on Earth?

More pertinent to your question is the fact that there are always an infinite set of lines that go through any finite set of points (the scientist's observations and data). So this is how experts disagree: they see the same points, but they've drawn a different line through them.


I was actually thinking about this question the other day - I haven't read all the past answers so I may be going over something that has already been covered.

For example in science, why is there disagreement when given access to the exact same information? Examples: Scientists disagreeing over string theory

Science has a two-fold purpose: 1) to provide means of finding new information, and 2) a social institution to spread that information, to convince other people that it is true.

As a social tool, it is inherently flawed, but it is also the best thing we've got to serve that purpose. It's flawed because as a social system it relies on people and has all the flaws of the people that make it up. Bias, corruption, and so on. But regardless of those things, it has proven itself to produce useful bits of knowledge and increase our understanding of the world and our ability to manipulate it.

When we talk of the question of how can physicists disagree about string theory, I think it's important to recognize the iceberg we're looking at with that question. You're looking at string theory as this important disagreement that makes science itself look foolish -- if science makes sense and is systematic, then how are they disagreeing?

But let's look under the water at the rest of this iceberg: all physicists who are in the debate about string theory pretty much unanimously agree on hundreds of years of physics research. On Earth, objects fall at about 9m/s/s. Gravity explains why we're bound to the planet, and why the planets revolve around the sun, and why the moon revolves around the earth. Relativity is true and has been proven by producing absolutely amazing predictions many decades ahead of time - we're still finding new evidence of Relativity today that was predicted ages ago!

Scientists have used science to come to a near-unanimous agreement on information that fills entire libraries. You are looking at the frontier and asking why is it so wild? It's wild because it's the frontier!


Let's start (by way of explanation) with a simple game. I will give a few numbers, and the game is to guess the next number in the series. So, I give the numbers 1, 2, 3...:

  • Anne says the next number is 4, thinking that I'm listing out the integers
  • Bob says the next number is 5, assuming that I'm thinking about Fibonacci numbers
  • Carmen says the next number is 10, because she's decided I've created a series using f(x) = x2 + 1 (for odd x) | x2 - 1 (for even x)

Who is correct? What if I said the next number was actually 6 (1, 2, 3, 6...), and asked you to predict the next number after that?

For any finite sequence of numbers there are an infinite number of functions that map to those numbers. Some of those functions are simple and straightforward, some are maniacal and crazed, and as a rule we prefer simple functions over crazed ones. But the crazed ones aren't wrong; they're merely unpalatable.

Likewise, for any finite number of empirical observations there are an infinite number of causal explanations which fit that data. Some are crazed, some are stupid, some are senseless, and we try to avoid those. But it is often the case that reasonable people will have reasonable disagreements about which of the more sensible models best fits that empirical data.

This is made more complicated by the fact that people don't always agree on what criteria should be used to group data points. If there are N observations, and observations n0 to nN-5 fit theory A, while observations n5 to nN fit theory B, which theory do we choose? And this becomes more fraught as we shift away from relatively simple topics like math and physics to philosophies and humanities. I mean, should Cézanne and Seurat really be considered post-impressionists, or are they better cast as 'late' impressionists? There is lots of room for civilized disagreement and scholarly debate on points like that.

The upshot is that we simple don't know what the 'truth' is — there are no 'facts' present in the world, just isolated experiences and theories/models we use to tie them together — and so there is always some room to re-model the world in a way we think makes more sense. we just need to be aware and wary of people want to re-model the world in ways that make less sense, for purposes of their own.


You examples are wildly divergent. The quality of a painting is purely subjective. My understanding is that string theory is largely about interpretation, rather than about expectation, and even less about experimental results. This is different from disagreeing about factual aspects of global warming (although even within just "global warming", there's a wide variety of types of disagreements; disagreement about historical data is different from disagreement over the cause of warming, and both are different from disagreement over predictions of future climate change).

Even when it comes to factual disagreement, no two people have exactly the same set of facts. Different people have read different articles, they have different experiences with the people involved with publishing those articles, and they have different life experiences. They also have different brains, that interpret data differently.

There is something called Aumann's agreement theorem that says that under certain conditions, people will not continue to disagree, but those conditions are not achievable in practice: both people have to have the same priors, and they have to have complete confidence in each others' reasoning, and complete confidence in each others' confidence in each others' reasoning, etc. Not only is the presence of irrational people enough to destroy the premises needed for the theorem, the possibility of irrational people is enough. In fact, the possibility that other people think there is a possibility is enough. Or the possibility that other people think that other people think there is a possibility, etc.


Maybe this is what happens in our lives daily.

We have an idea to spend the weekend, then our wife and children enrich the concept and the weekend we were thinking would great... it becomes FANTASTIC.

We have a great idea on a financial structure to acquire and develop a company.After we discuss it... it becomes FANTASTIC.

Why does not the above happens more often ? Because of our nature. We are complex creatures with a lot of bias which play a role in our decisions at the conscious and unconscious level.

Why was Martin Heidegger Nazi and Hannah Arendt was not Nazi ? They had the same facts: stupid people with power is still stupid people, for example.

Why do the scientists who are funded directly and indirectly by fossil fuel companies do not transform their job in favor of postponing or even eliminating the calamity which is approaching at high speed ? Well the answer must be related with money, status, perks, sex...

These people are in charge of funding.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .