Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Should metaphors be used in Science, or is it the logical fallacy of metaphor?

  1. Aug 23, 2009 #1
    Okay, here's some food for thought. You hear people talk about the "logical fallacy of using metaphors" to prove a point, and that evidence rather than metaphors should be used to prove.

    However, at the same time metaphors can help get meaning across where literal language struggles and might be confusing.

    So how do you conceptualize all this as far as communication goes? If you want to be as honest/integrity as possible, but also be a decent communicator and not have people lost?
  2. jcsd
  3. Aug 23, 2009 #2
    metaphors aren't used to prove a point, they are used to explain a point.
  4. Aug 23, 2009 #3
    For example, original scientific peer-review journal articles don't use metaphors. So I'll read several peer-review articles, brainstorm some follow-up hypotheses (for fun). I'll use my knowledge to look for possible details which "may not fit" and brainstorm alternative explanations/principles for what's in the peer-review article. Or I'll find other ways to brainstorm new ideas.

    Then I'll share them with my roommates. They act really confused. So I want to make it so they can understand me much easier, and that I'm asking for input in the first place (I don't think they understand that part). Then at the same time I don't want to accidentally mislead the wrong way with a metaphor. I want to be as honest in science as possible. Although I'm not Albert Einstein, wasn't he that way?

    So that's part of the reason I'm pondering this, among others.
  5. Aug 23, 2009 #4


    User Avatar
    Gold Member

    I think that anything goes as long as it gets a basic idea across. Once that is absorbed, one can gradually elaborate upon the explanation until it is educationally correct. After all, you can't explain scattering to a 4-year-old when he asks why the sky is blue, but you also don't want to discourage his interest. I'd say something like "the air only likes blue, so it sends the other colours away". It's not exactly something that you'd expect to read in 'Nature' magazine, but it works.
  6. Aug 24, 2009 #5
    "Science is like an old friend, sometimes it tells you things you don't want to hear." - C. Darwin

    Well, that is actually a simile, but you get the point.. maybe.. I'm not even sure I get the point.

    Ok, here is a link http://www.morphostasis.org.uk/metaphors.htm [Broken] that I did not read. Let me know if it's any good.
    Last edited by a moderator: May 4, 2017
  7. Aug 24, 2009 #6


    User Avatar
    Homework Helper

    <rant> I don't like the idea that science should be restricted. When someone asks about whether the scientific method, or falsifiability, or reproducibility underlies science, I cringe. Everybody knows what science is; everybody knows the importance of evidence, the possibility of flukes, and the need for testing ideas to make sure they're right. The scientific method is a fancy way of saying "common sense".</rant>

    So, metaphors? Sure, if they're used correctly and honestly. They can of course be abused, but so can everything.
  8. Aug 24, 2009 #7


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I think you've already answered your own question. One wouldn't use metaphor (or analogy...I'm not sure if that's what you really mean here given the context) as a step in a proof or as part of their hypothesis or experimental design. On the other hand, if it aids in communication to explain a complex concept to someone for the first time, that's communication, not scientific method. The only issue of "integrity" is that you should make sure the person is aware you are using a metaphor or analogy to explain a concept to them in the event they are sufficiently naive on the subject not to realize it for themselves, and so they understand that it may be an imperfect explanation to help get them started, and that there will be more details they can fill in later when they have more background to better understand it.
  9. Aug 24, 2009 #8
    Yes, I think you might be thinking of analogies. But, metaphors are useful, and are used. Think of scattering cross-sections. The common unit developed right away was..."Wow, that thing's as big as a barn." So, we have the barn. That's a metaphor. It works.
  10. Aug 24, 2009 #9
    Jeez, this is bright as an inversebarn in here !
  11. Aug 24, 2009 #10
  12. Aug 24, 2009 #11
    So then what's your input on the subject? Do you think a metaphor could make Science "bright"?
    Last edited: Aug 24, 2009
  13. Aug 24, 2009 #12
    I'd better refrain from silly comments since brightness is more accurately described as inversebarn per unit of time. But since you request a serious comment :

    An analogy or a metaphor definitely has value, provided it is used in a precise scientific context. If the analogy is used to illustrate an equation, it's fine. If it replaces the use of equations, it's moot.
  14. Aug 25, 2009 #13
    Thought experiments are often used to explain ideas. They may be more appropriate in formal science than metaphors.
  15. Aug 25, 2009 #14
    Formal Science? What about scientific areas outside of peer-review journals? Experts reading the peer-review journal often understand context really well, so they don't need extra help. However, in having conversations between scientific disciplines or those without as much science education?

    For example, if I say to someone, "Have you noticed that some have trouble with group conversations and others don't? Some keep on trying to speak up and always get interrupted, or are accused of interrupting, or just are very quiet because they've had trouble in the past? Well, just like people throwing a ball back and forth, there's research that people use their body language and eye contact to pass the conversation back and forth. Many of those who have trouble cutting into the conversation on time, or are always talked over may not grasp this language." If I say it that way, someone may say "That's interesting". However, if I leave out the metaphor of a ball being tossed back and forth, or "have you noticed this", then people act confused and say "you shouldn't believe every study you read". However, peer-review journals leave that stuff out. So I want to be as honest/objective in thinking about science with roommates, but at the same time not confusing them. I just don't want to feel like I'm leading them on. I want them to think for themselves, but not act like someone who's ignorant to science.

    If that makes sense?
  16. Aug 25, 2009 #15
    Or another food for thought, people often use the phrase, "You can't think out of a vacuum. You've gotta have something." If I use what I know from neurology and cognitive science and blend them together to invent the equivalent of "you can't think from a vacuum" right on the spot, and brainstorm experiments to test it, because I want to find a new/different way of thinking about things, my roommates act very confused. (Their look :confused:) Giving much scientific background knowledge will just confuse them even more and make them think it's a monologue, rather than some scientific speculation I just brainstormed right on the spot and am looking for feedback on.

    I wonder if using metaphors and figures of speech that they're familiar with may help get the point across, then I could bring in any relevant scientific evidence and then how my scientific speculations come into play?

    Although none of us are Einstein, just food for thought: He was a complex/great thinker, :smile: but his speeches confused most everyone and he would speak for hours on end; most of his audiences didn't even need any bedtime stories. :eek: So, is it possible to be extremely scientific and honest, but also make sense and actually be interesting, to those who don't have background knowledge or even care about Science?

    This is what I'm very curious about.
  17. Aug 25, 2009 #16


    User Avatar
    Gold Member

    Ee-yikes! This is exactly wrong.

    The people who have the most difficulty with science are the ones who do not understand reproducibility and falsifiabilility.


    How can you say such a thing!??
    "Common sense" is precisely the thing that people use to refute science. Science is exactly the opposite of common sense. "Common sense" is simply the self-fulfilment of what you think you already know.

    Example: it was "common sense" in the pre-Galilean world to "know" that a watermelon (which weighs twenty times what an orange does) will fall at twenty times the speed. It wasn't until Galileo made his climb of the Tower of Pisa that "common sense" was shown to be horribly, horribly bad at getting us to learn anything new.

    "Common sense" argument is one of the primary weapons that fundies sue to manipulate their followers.

  18. Aug 25, 2009 #17


    User Avatar
    Gold Member

    Junglebeast's comment is worth repeating:

    "metaphors aren't used to prove a point, they are used to explain a point."

    Your metaphor of a ball being passed back and forth helps explain the phenomenon. But it's not part of the science of the phenomenon.

    The problem with metaphors is that, by definition, they are inaccurate. (If they weren't inaccurate, they wouldn't be a metaphor) They will always break down at some level of scrutiny, and the temptation is great to carry a metaphor too far. For example: 'what if I keep the ball and don't pass it on?' This does not map onto the dynamics of a conversation: there is no way of forcing others to wait until you give up your turn at talking. The passing ball metaphor breaks down.

    The metaphor helps explain some of the science, but you cannot use it to extrapolate.

    It is not possible to be perfectly scientific and be metaphorical at the same time. However, it is possible to switch back and forth between the two in the course of a single explanation.
    Last edited: Aug 25, 2009
  19. Aug 25, 2009 #18
    You can certainly use metaphors to help laymen understand your thoughts and ideas. As Dave points out though the metaphor could wind up giving your audience the wrong impression.

    I think that he's refering to the scientific method as being common sense, not necessarily the evidence or conclusions. What most people who have little interest or education in science probably have a problem with is that common sense rules or the scientific method come up with things that seem to contradict common sense. Their initial reaction is likely to think that there is something wrong with the way that you went about what you were doing. Which would actually support ideasrule's idea that scientific method is common sense. Obviously if you get unexpected results you need to test for repeatability and make sure that your procedures are not flawed.

    I don't think that the pseudoscientists use common sense so much as they use overly simplified arguments and explinations that they dress up as common sense. And of course there is always the saying that common sense isn't so common.

  20. Aug 25, 2009 #19


    User Avatar
    Homework Helper

    Try telling those people that you have an undetectable dragon in your garage and see what they say. Ask them whether Russell's teapot exists.They may fiercely argue that their particular religion doesn't need to be falsifiable, but that doesn't mean falsifiability isn't a notion they respect and cherish.

    For reproducibility: ask somebody that if 10 experiments prove A is false and 1 dubious experiment seems to prove A is true, whether A is true or not. They may say that more experiments are needed, or that the one contradictory result should be investigated, or that there's enough evidence contradicting A to accept it as false, but hardly anybody will say A is the absolute, undeniable truth.

    I meant the scientific method was common sense, not any preconceived notions. I thought that was clear. If I want to see how well a car performs, hardly anybody will object to using the "scientific method" and taking it for a test drive. If I want to claim that my enemy murdered my mom, nobody will claim that I don't need lots of objective evidence and down-to-earth explanations of why they prove my case.

    People only object when common-sense concepts like experimentation and logic are applied to the natural world. They have a double standard: the scientific method works perfectly for everyday affairs, but nature is somehow magical, so they resort to making things up. (Alternatively: the scientific method works perfectly for everyday affairs, but it contradicts religion, so it can't be used to study nature.)
  21. Aug 25, 2009 #20
    I have not understood what you are conveying here. I would appreciate if you can elaborate for me. The way I see it, there is simply no way an experiment proves anything true, therefore already asking the question "whether A is true or not", independently of any experiment, is a misunderstanding of the scientific method. At best, a hypothesis might not be falsified (yet), or at strongest, we may make some form of likelihood agreement between the hypothesis and the data. However there is never any "absolute" likelihood, in the sense that we cannot define a measure (or probability) in the "space of hypothesis" which would allow us to quantify the likelihood of the hypothesis given the data (unless you are conveying extreme (unreasonable) bayesian inference where the available data can never be biased since, by definition, all the available data is what defines reality).
  22. Aug 26, 2009 #21

    Math Is Hard

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    My personal experience with college roommates is that only 50% were worth interacting with socially. I never had a driving need to explain to any of them what I was learning. If they didn't have a desire to think for themselves or understand things deeply, I really couldn't give a rat's patoot. It may sound nasty, but I just saw them as the chumps that kept me at the top of the curve. (Thanks, suckers. :smile:)

    In the few cases that I did care to explain something to someone who asked:

    1) I did my best, usually starting with metaphor/s (you might need to work up a few to find the right one). And as a cautious undergrad, I'd usually offer the caveat of "this analogy is how it makes sense to me".
    2) If requested for more detail, I'd explain at intermediate detail to the best of my ability, and acknowledge that I was just a beginner and might be giving a less than perfect explanation
    3) If requested for even more detail, I'd usually ask the person to consult an expert or a textbook or some other reputable source that I could provide.

    That's about all I have to say, other than I don't think this thread has much to do with logical fallacies; it's really just a question of pedagogy and what's appropriate in what situations.
    Last edited: Aug 26, 2009
  23. Aug 26, 2009 #22
    Thanks for the the advice you gave. My problem is what analogies make sense to me are going to be extremely different than what will work for them. One man's detail is another man's concept, and vice versa.

    By logical fallacy I was more meaning that I want to make sure I avoid the logical fallacy of proving by metaphor, while balancing helping them understand. So I guess how Science, outside of peer-review journals, can combine pedagogy with the avoidance of logical fallacies.
  24. Aug 26, 2009 #23


    User Avatar
    Gold Member

    Have you seen a show called Big Bang Theory? It might help you understand your difficulty communicating with your roommates. :wink:
  25. Aug 27, 2009 #24
    Hmmm, I'll have to watch that.

    One thought I had, although I think it's more of making sure it's worded honestly than making sure there's no possible way for being accidentally mislead (they don't use layman roommate explanations in peer-review journals): I could start with the part of trying to relate the scientific concept to them and a metaphor, then say something like, "Science has 'actually tested' this idea" and introduce it in the scientific manner. (Since in Science they say it doesn't matter how you get there, but instead whether it's tested in the end)

    Although any roommate at random may not possibly be 100% objective in thinking about the peer-review journal articles, at least that way I think I'd be completely avoiding any proof by metaphor. I'd be separating the relating to them in their own words part from the actual scientific evidence part.

    Hmmm, maybe I'm over thinking this? But honesty/objectivity is something important to me, while at the same time not having them confused. :smile:
  26. Aug 27, 2009 #25


    User Avatar
    Gold Member

    You must cater your explanation to your audience. Again, it is impossible to be both scientifically accurate and at the same time, use simplifying metaphors. A metaphor, by definition, is not an accurate description of something.

    You cannot describe an atom as a little moon orbiting a planet and be accurate in doing so.
    Last edited: Aug 28, 2009
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook