RealClimate logo

On arguing by analogy

Filed under: — gavin @ 2 September 2014

Climate blogs and comment threads are full of ‘arguments by analogy’. Depending on what ‘side’ one is on, climate science is either like evolution/heliocentrism/quantum physics/relativity or eugenics/phrenology/Ptolemaic cosmology/phlogiston. Climate contrarians are either like flat-earthers/birthers/moon-landing hoaxers/vaccine-autism linkers or Galileo/stomach ulcer-Helicobacter proponents/Wegener/Copernicus. Episodes of clear misconduct or dysfunction in other spheres of life are closely parsed only to find clubs with which to beat an opponent. Etc. Etc.

While the users of these ‘arguments’ often assume that they are persuasive or illuminating, the only thing that is revealed is how the proposer feels about climate science. If they think it is generally on the right track, the appropriate analogy is some consensus that has been validated many times and the critics are foolish stuck-in-the-muds or corporate disinformers, and if they don’t, the analogy is to a consensus that was overturned and where the critics are the noble paradigm-shifting ‘heretics’. This is far closer to wishful thinking than actual thinking, but it does occasionally signal clearly who is not worth talking to. For instance, an article pretending to serious discussion on climate that starts with a treatise about Lysenkoism in the Soviet Union is not to be taken seriously.

Since the truth of falsity of any scientific claim can only be evaluated on it’s own terms – and not via its association with other ideas or the character of its proponents – this kind of argument is only rhetorical. It gets no-one closer to the truth of any particular matter. The fact is that many, many times, mainstream science has survived multiple challenges by ‘sceptics’, and that sometimes (though not at all often), a broad consensus has been overturned. But knowing which case is which in any particular issue simply by looking for points of analogy with previous issues, but without actually examining the data and theory directly, is impossible. The point being that arguments by analogy are not persuasive to anyone who doesn’t already agree with you on the substance.

Given the rarity of a consensus-overturning event, the only sensible prior is to assume that a consensus is probably valid absent very strong evidence to the contrary, which is incidentally the position adopted by the arch-sceptic Bertrand Russell. The contrary assumption implies there are no a priori reasons to think any scientific body of work is credible which, while consistent, is not one that I have ever found anyone professing in practice. Far more common is a selective rejection of science dependent on other reasons and that is not a coherent philosophical position at all.

Analogies do have their place of course – usually to demonstrate that a supposedly logical point falls down completely when applied to a different (but analogous) case. For instance, an implicit claim that all correct scientific theories are supported by a unanimity of Nobel Prize winners/members of the National Academies, is easily dismissed by reference to Kary Mullis or Peter Duesberg. A claim that CO2 can’t possibly have a significant effect solely because of its small atmospheric mixing ratio, can be refuted as a general claim by reference to other substances (such as arsenic, plutonium or Vitamin C) whose large effects due to small concentrations are well known. Or if a claim is made that all sciences except climate science are devoid of uncertainty, this is refuted by reference to, well, any other scientific field.

To be sure, I am not criticising the use of metaphor in a more general sense. Metaphors that use blankets to explaining how the greenhouse effect works, income and spending in your bank account to stand in for the carbon cycle, what the wobbles in the Earth’s orbit look like if the planet was your head, or conceptualizing the geologic timescale by compressing it to a day, for instance, all serve useful pedagogic roles. The crucial difference is that these mappings don’t come dripping with over-extended value judgements.

Another justification for the kind of analogy I’m objecting to is that it is simply for amusement: “Of course, I’m not really comparing my opponents to child molesters/food adulterers/mass-murderers – why can’t you take a joke?”. However, if you need to point out to someone that a joke (for adults at least) needs to have more substance than just calling someone a poopyhead, it is probably not worth the bother.

It would be nice to have a moratorium on all such analogical arguments, though obviously that is unlikely to happen. The comment thread here can assess this issue directly, but most such arguments on other threads are ruthlessly condemned to the bore-hole (where indeed many of them already co-exist). But perhaps we can put some pressure on users of these fallacies by pointing to this post and then refusing to engage further until someone actually has something substantive to offer. It may be pointless, but we can at least try.

210 Responses to “On arguing by analogy”

  1. 201
    jgnfld says:

    @178 @188 The argument used here assumes that the MMPI scales are independent. They are not, and in fact studies of the underlying factor structure of the MMPI-2 items end up with more like 5 statistically independent factors being measured.

    There are reasons for going the way they did which are not germane here.

  2. 202

    “Appreciating this would negate the need to grasp around for reasons why the majority are not with ‘us’ but with ‘them’.”

    – See more at:

    The central assumption of that sentence is quite simply untrue:

    • Americans are more than two times more likely to vote for a congressional or presidential candidate who strongly supports action to reduce global warming. Democrats, liberal and moderate Republicans, and Independents are more likely to vote for such a candidate. Only conservative Republicans are less likely to vote for such a candidate.

    • Likewise, Americans are three times more likely to vote against a political candidate who strongly opposes action to reduce global warming. Only conservative Republicans are, on balance, slightly more likely to vote for a candidate who strongly opposes action to reduce global warming.

    Various other measures of opinion clearly show that the “majority is with us, not them.” And the result of this poll is consistent with a long string of such research. Unfortunately, that majority is not yet sufficiently exercised on the topic–climate change should have more like a 10x voting factor. Deniers make more noise than their numbers merit.

  3. 203
    Dan H. says:

    I was referring to scientists, not the general public, or worse, politicians. The latter groups are much more likely to engage in ad hominem attacks, trying to portray their opponents as inferior, and by extention, and work they may present. Scientists will criticize the research, not the scientists.

  4. 204
    Roscoe Shaw says:

    @200 I misread. Sorry. Tried again and still don’t really get it. But I’ve always been better at science than psychology or debate.

  5. 205
    Edward Greisch says:

    Gavin’s ‘On Arguing by Analogy’ is soft on them compared to “A Manual for Creating Atheists” by Peter Boghossian Chapter 8. You should read what Boghossian has to say about multiculturalism, relativism, faith based reasoning [not necessarily in the religious sense], etcetera.

    “Cognitive, epistemological, and moral relativism are toxins that students trained in the humanities regularly consume in large doses.”

  6. 206
    Hank Roberts says:

    OK, I tried another analogy, over at

    to answer why we show climate sensitivity (for this current planet and condition) as a range of variously likely numbers not closing in on a single precise number.

    Pointers to a better fifth-grade-level explanation always welcome.

  7. 207
    Sean says:

    How Brains Think, 1 hr cognitive science lecture May 2014

    Viewing up to the 7 minute mark will be enough to know if it’s relevant to climate discussions and possibly useful for the viewer who’d like to learn something new.

  8. 208
    Edward Greisch says:

    My letter to the editor before reading “A Manual for Creating Atheists” by Peter Boghossian:

    On the subject of Global Warming, Cheri Bustos and Bobby Schilling gave VoteSmart different answers. The question is: “Environment: Do you support the federal regulation of greenhouse gas emissions?”

    Cheri Bustos: Unknown Position

    Bobby Schilling: No

    Since Global Warming will shut down the food supply right here in the USA in something like 40 years, Bobby Schilling is showing himself to be anti-farmer and anti-grocery-business. Bobby Schilling must be a psychopath to put immediate profits ahead of his children’s and grandchildren’s lives.

    We know from the history of previous civilizations that have crashed, that when there is no food in the grocery store, people drop their tools wherever they are and wander away looking for food, which they almost never find.

    Cheri Bustos’s non answer is better because the question is a science question, not a political question. That is a reason to vote for Cheri Bustos.

    The thing you can do to solve Global Warming while keeping the government out of it is to sell your stock in coal mines. Or, if the stock holders put coal out of business without selling the mine, then Global Warming will slow down a lot without government intervention.


    My letter to the editor after reading “A Manual for Creating Atheists” by Peter Boghossian:

    On the subject of Global Warming, Cheri Bustos and Bobby Schilling gave VoteSmart different answers. The question is: “Environment: Do you support the federal regulation of greenhouse gas emissions?”

    Cheri Bustos: Unknown Position

    Bobby Schilling: No

    You see 2 “experts” who say contradictory things or maybe a scientist high in the government who contradicts the 97% majority of working climate scientists. How do you know who is right? You need to know some science yourself.

    Science is a method of finding out for yourself, not a belief system. The crucial experiment was first done in the year 1859. Carbon dioxide blocks infrared, preventing heat from escaping Earth into the super-cold of space.

    Cheri Bustos’ position is better than Bobby Schilling’s because Bobby Schilling is pretending to know something that Bobby Schilling doesn’t understand.


    Boghossian says: “Attack the faith, not the religion.” “Faith is unreliable epistemology.” “Faith means pretending to believe something you don’t know.”

    Denialists are “precontemplative.” They are “doxastically closed.” They are not trying to understand because they think they do understand. Since they don’t understand climate science, they make up absurd motivations such as “world domination.”

    Boghossian says: “Attack the faith, not the religion.” “Faith is unreliable epistemology.” “Faith means pretending to believe something you don’t know.” Don’t argue, especially facts. Ask them questions about their faith in the passive voice.

    I am reading the book for the second time. Rather than me trying to repeat the book to you, it wold be better for you to read the book.

  9. 209
    patrick says:

    Werner Heisenberg, he says, always looked for a good picture or analogy, as did Richard Feynman, who invented “a sort of pictorial calculus.” Even with an apt analogy in hand, Schellnhuber admits that with many leaders, “you cannot give a scientific lecture. But you try to do it in terms that can be felt intuitively, perceived by people. Sometimes you succeed and sometimes you fail.”

    –from the profile of H.J. (John) Schellnhuber in PNAS:

  10. 210
    Russ Doty says:

    A response to a naysayer who wondered why anyone would “continue to push climate alarm-ism” was: “It’s not because of how we make money. It’s because we know man-caused global warming is harmful. Just ask the Pope, other Catholics, Evangelicals including Rick Warren, World Council of Churches (, and other denominations. To see if your denomination supports eco-justice (or what climate change naysayers call “alarm-ism”), Google your denomination (Methodist, etc.) and “climate change.””