Science as an intellectual “safe space”? How to do it right.

I don’t recall hearing the term “safe space” until recently, but now it seems to be used all the time, by both the left and the right, to describe an environment where people can feel free to express opinions that might be unpopular in a larger community, without fear of criticism or contradiction.

Sometimes a safe space is taken to be a good thing—a sort of hothouse garden in which ideas can be explored and allowed to grow rapidly in an environment free of natural enemies or competition—and other times it’s taken to be a bad thing, a place where ideas will not develop in useful ways.

The short version is that people sometimes (but not always) want safe spaces for themselves, but they typically are derisive of safe spaces for people they disagree with. Safe spaces are a little like protective tariffs in economics: if you feel strong, you might not see the need for anyone to have a safe space; if you feel weak, or if you have weak allies, you might want those protective zones.

Psychology journals as safe spaces

This all came up when I was thinking what seemed to me to be exaggeratedly defensive reactions of scientists (in particular, some psychology researchers) to criticism of their published work. To me, if you publish your work, it’s public, and you should welcome criticism and active engagement with your ideas. But, to many researchers, it seems that praise is OK but skepticism is unwelcome. And in some cases researchers go all out and attack their critics. What these researchers seem to be looking for is a safe space. But to me it seems ridiculous to publish a paper in a public journal, promote it all over the public news media, and then claim object to criticism. This sort of behavior seems roughly equivalent to fencing off some area in a public park and then declaring it private property and only admitting your friends. Actually, even worse than that, because prominent psychologists use their safe space to spread lies about people. So it’s more like fencing off some area in a public park and then declaring it private property and then using it to launch missiles against people who you perceive as threatening your livelihood.

“Freakonomics” as a safe space

Another example came up a few years ago when Kaiser Fung and I wrote an article expressing a mix of positive and negative attitudes toward the Freakonomics franchise. One of the Freaknomics authors responded aggressively to us, and in retrospect I think he wanted Freakonomics to be a sort of safe space for economic thinking. The idea, perhaps, was that “thinking like an economist” is counterintuitive and sometimes unpopular, and so it’s a bad idea for outsiders such as Kaiser and me to go in and criticize. If the Freaknomics team are correct in their general themes (the importance of incentives, the importance of thinking like an economist, etc.), then we’re being counterproductive to zoom in on details they may have gotten wrong.

Safe spaces in science

I have no problem with my work being criticized; indeed, I see it as a key benefit of publication that more people can see what I’ve done and find problems with it.

That said, I understand the need for safe spaces. Just for example, I don’t share the first draft of everything I write. Or, to step back even further, suppose I’m working on a math problem and I want to either prove statement X or find a counterexample. Then it can be helpful to break the task in two, and separately try to find the proof or find the counterexample. When working on the proof, you act as if you know that X is true, and when searching for the counterexample, you act as if you know X is false. Another example is group problem solving, where it’s said to be helpful to have a “brainstorming session” in which people throw ideas on the table without expectation or fear of criticism. At some point you want to hear about longshot ideas, and it can be good to have some sort of safe space where these speculations can be shared without being immediately crushed.

My suggestion

So here’s my proposal. If you want a safe space for your speculations, fine: Just label what you’re doing as speculation, not finished work, and if NPR or anyone else interviews you about it, please be clear that you’re uncertain about these ideas and, as far as you’re concerned, these ideas remain in a no-criticism zone, a safe space where they can be explored without concern that people will take them too seriously.

34 thoughts on “Science as an intellectual “safe space”? How to do it right.

    • I suspect the reason is that publications are not often viewed as a method of sharing ideas, but rather a mechanism towards career advancement. Criticism threatens that.

      • “I suspect the reason is that publications are not often viewed as a method of sharing ideas, but rather a mechanism towards career advancement. Criticism threatens that.”

        Or publications can be used for reaching many other (possibly related) goals concerning politics, influence, money, etc. Criticism could threaten all of those goals as well i reason.

        I actually think a large part of people in academia today might actually act, and talk, more like lawyers, politicians, managers, or sales people.

        I think this might stem from decades of unscientific processes that may have gone on at institutions (e.g. see Binswanger: https://link.springer.com/chapter/10.1007/978-3-319-00026-8_3).

        The results of this can perhaps be seen in things mentioned in this blog post: people who call themselves scientists act very unscientific (and do not even seem to notice, and or mind, that). And things like “safe places” are apparently necessary in places that should above all value, and be about, reasoning, the sharing of ideas, and free speech.

        I think Binswanger may have been correct when he wrote the following:

        “The perverse incentives created by the peer-review process ensure that the steadily increasing number of published articles in scientific journals often does not lead to new or original insights and, therefore, many new ideas do not show up in established journals. They can rather be found in books and working papers, where there is no pseudo-quality control which hinders innovative ideas.”

        I sometimes wonder whether the same can be said concerning current academics: just like actual useful, smart, and original insights are possibly mostly found outside scientific journals, perhaps actual useful, smart, and original scientists are possibly mostly found outside universities/academia…

    • Certain ideas will be immediately rejected by certain groups of people, which may not be the best feedback. Additionally, criticism that is too harsh may disincentive people from putting forth further ideas. Of course, taking the focus away from negative feedback doesn’t necessarily mean everyone will be a yes man.

    • To generalize what Garnett and Bobbo say: One point of doing so is to signal membership or opposition to a certain group. Another is to allow others to signal back the nature of your membership in a group.

  1. “Just label what you’re doing as speculation, not finished work, and if NPR or anyone else interviews you about it, please be clear that you’re uncertain about these ideas and, as far as you’re concerned, these ideas remain in a no-criticism zone, a safe space where they can be explored without concern that people will take them too seriously.”

    But if you don’t want people to take the ideas too seriously, why would you be giving interviews about them on NPR? And if you consider your work or ideas to be speculative, why would you not want criticism: how will the ideas grow and improve without it? It seems to me that the kind of “safe space” that has developed in some areas, and is very problematic, is one in which the authors do want their ideas taken seriously. In fact, they want their ideas taken as gospel truth.

  2. I think the importance of (adherence to) scientific values and principles like proper rules of debate are crucial in science.

    I fear however, that this might be very hard to actually follow for a lot of folks currently in academia, even those that think they are on, or part of, a possible or imaginary “good” side.

    I would like to add that it’s not enough to say things like “i value criticism” and then not actually do anything with it when you receive it, or even worse, reply to that criticism in an unscientific manner.

    You can’t say one thing, and do another. And you can’t pick and choose when, and with whom, your scientific values and principles apply and when not.

    Or to quote Bruce Lee:

    “Knowing is not enough, we must apply
    Willing is not enough, we must do”

  3. Consider the word “peer” in the phrase “peer review”. Now consider the trend in certain fields for journal editors to request that authors nominate a list of potential reviewers. If you like, think of that as a list of those the author feels are his or her peers.

    In that setting is it any surprise that, having cleared the hurdle of peer review, some authors try to assert a continuing right to choose between allowable and non-allowable sources of criticism?

    In my opinion, this is just another manifestation of the tribalism that infests every aspect of our culture.

    • “Consider the word “peer” in the phrase “peer review”. Now consider the trend in certain fields for journal editors to request that authors nominate a list of potential reviewers. (…) In my opinion, this is just another manifestation of the tribalism that infests every aspect of our culture.”

      Yep, it’s quite astonishing, and shameful, what they have done, and are currently doing, to science.

      Should you be interested, here’s a paper that mentions the problematic nature and possible origin of many things, including the journal-editor-peer-review system: https://link.springer.com/chapter/10.1007/978-3-319-00026-8_3

      What i find most astonishing though, is that many of the proposed “improvements” to “fix” science may actually reinforce the exact problematic system, and roles, that got science in the mess it is in. All kinds of “special” tasks, “special editors”, and roles for peer-reviewers, and even entirely new 3rd parties, have already been, and currently are being thought of to “improve” matters. To me, it’s like they read Binswanger’s paper linked to above, and used it as a playbook for all their efforts to improve” matters. It’s astonishing to me.

  4. You are using a bit of an atypical definition of “safe space” here as you assert that it is an “environment where people can feel free to express opinions that might be unpopular in a larger community.” Traditionally, “safe spaces” are intended to be safe for PEOPLE, not for ideas. The concept of a safe space came about from spaces where certain types of discourse were barred because they harmed marginalized participants and thus kept them from participating. In the US I believe this first arose around feminism and later expanded into the gay community.

    The reason that homophobic ideas are barred from discussion in some”safe spaces” isn’t that folks want to have a conversation without alternative viewpoints, it is because allowing homophobia into the space would prevent gay folks from being able to participate fully.

    Source: https://en.wikipedia.org/wiki/Safe_space

    • Quote form above: “Traditionally, “safe spaces” are intended to be safe for PEOPLE, not for ideas”

      In the wikipedia page you link to the following is mentioned: “The terms safe space (or safe-space), safer space, and positive space may also indicate that a teacher, educational institution or student body does not tolerate violence, harassment, hate speech or dissenting views, thereby creating a safe place for all people.”

      I am all for safe spaces that do not tolerate violence, harassment, and hate speech. I am however not sure how to interpret the “dissenting views” part of the above description of “safe places”.

      In my interpretation, “dissenting views” should be welcomed in science and scientific discourse (as long as they are not hate speech, tolerate violence, etc).

      If this makes any sense, isn’t the description of a “safe place” in the wikipedia link you provided an example that “safe places” could be (or could become?) about barring ideas and/or alternative viewpoints?

      • I use to post to one message board that harbored some of the crankiest and even nastiest posters who are practiced in the art of emotional blackmail. So I welcome safe spaces more now. On the other hand, I do miss, occasionally, giving such individuals a run for their money by outdebating them. On Twitter, however, I have yet to come across a nasty responder. I’m puzzled why I haven’t yet had any major nasty experience

        In addition, I cast most of my opinions as ‘speculations’. It does spare me some grief I think. I also don’t bill myself as an authority either, which is liberating.

        • Quote from above: “I use to post to one message board that harbored some of the crankiest and even nastiest posters who are practiced in the art of emotional blackmail.”

          Yes, things can get nasty. That’s why i think it’s very important that everyone keeps reminding, and correcting, everyone else when someone uses unscientific tactics, or language. I however think this will (sadly) be dependent on the specific forum where the debate takes place.

          I myself don’t like to read twitter anymore, because i just can’t deal with (what i view as being) unscientific tactics, and communication. I’ve seen “scientists” mention on their twitter account that they should follow person X, or that they should follow person Y so they can get above X amount of followers. This all reminds me more of voting on “american idol” or of trying to be “an influencer” or something like that, than of anything scientific. Anyway, i don’t like most of (what i think the kids these days call) “social media”, especially not when it involves science.

          I worry though that real debate will become more and more impossible at universities, and even on other forums. I even think many scientists aren’t even capable of engaging in proper scientific debate anymore, partly because i think (logical) reasoning is not even being taught anymore. Combined with other problematic issues at universities, i think this may have resulted in what i call a debating style that more resembles that of a politician or lawyer, if it takes place at all.

          The entire thing also reminds me of an episode of “Family guy” which takes a shot at “millenials” (1.30 – 1.55 is the absolute highlight for me):

          https://www.youtube.com/watch?v=6CBeAnzNka4

      • In my experience, the “dissenting views” that are banned in safe spaces typically align well with the types of speech that you do not want included in scientific discourse (“hate speech, tolerate violence, etc”). I agree that this could be depicted as “barring ideas and/or alternative viewpoints,” but I think that is a deceptive way to depict the intent and impact of these spaces. In fact, the quote you pulled from Wikipedia concludes by saying that the purpose is “creating a safe place for all people.”

        Part of the reason that I made my comment was that Andrew had asserted that “safe spaces” exist to protect certain ideas. I find that this view is quite common among folks who rarely participate in “safe spaces.”

        • Quote from above: “In my experience, the “dissenting views” that are banned in safe spaces typically align well with the types of speech that you do not want included in scientific discourse (“hate speech, tolerate violence, etc”). I agree that this could be depicted as “barring ideas and/or alternative viewpoints,” but I think that is a deceptive way to depict the intent and impact of these spaces’

          I am not really familiar with “safe spaces”, so please forgive me if i am misunderstanding. I am simply reading the wikipage which you linked to.

          The quote from the wiki page clearly mentions what i think you mean with “the types of speech that you do not want included in scientific discourse (“hate speech, tolerate violence, etc”)” *before* using the term “dissenting”. I interpret this to imply that “dissenting” is something different from those things.

          This was further confirmed by simply clicking on the hyperlink of the word “dissenting” in the actual wiki page, and even looking up the word “dissenting”. I just did this all for the 2nd time, and i remain by my interpretation. This could possibly reflect a mistake in the wiki page, but i am simply reading what is says on the wiki page.

  5. Since the theme is criticism, as anon pointed, your usage of safe space seems a bit off. I’m no expert either, but it seems to me that safe space is about power.
    Men in a patriarchal society will abuse and attack woman who engage in discussion in a certain way. Thus the need to build safe spaces for them to engage in discussion without being abused or attacked.
    Again, is not about not being criticized or contradicted, but attacked or abused in a situation in which the attacker has much more power. It is a way to expand the space of speech, to prevent powerful people to attack the oppressed in their speech and discussion.

    • (i replied to your comment here https://statmodeling.stat.columbia.edu/2019/01/22/science-as-an-intellectual-safe-space/#comment-953818 but i messed up directly connecting it to your comment here. This is my 2nd attempt at doing that. I will copy the comment below here:)

      Quote from above: “Men in a patriarchal society will abuse and attack woman who engage in discussion in a certain way. Thus the need to build safe spaces for them to engage in discussion without being abused or attacked”

      This seems an incorrect statement, and/or a gross generalization.

      It could be that in certain forums, or at certain times, or in certain instances, a man could attack a woman in an unscientific manner, but this of course does not (need to) happen always (or merit the use of the word “will” in your quote).

      If, and when, that happens though it seems crucial for others to point to the possible unscientific, or even abusive, tactic or communication.

      Just like, others should point out when someone says, or writes, something that could be unscientific, untrue, or severely flawed.

      I reason, this will also help to make “safe spaces” less and less necessary, will help to keep everyone involved in the discussion (and not create “bubbles” or separate groups), and will help to keep scientific values and principles (also concerning debating) intact.

      • Just to clarify. My statement “men in a patriarchal society will…” is meant to be: one of the average causal effect of a patriarchal society is that men will attack women etc. In other words, it is not a deterministic statement, but a probabilistic one.

        • Quote from above: “Just to clarify. My statement “men in a patriarchal society will…” is meant to be: one of the average causal effect of a patriarchal society is that men will attack women etc. In other words, it is not a deterministic statement, but a probabilistic one.”

          Thank you for the explanation/nuance.

          Leaving aside whether there even is a “patriarchal” society, in let’s say the U.S.A., if a “patriarchal” society enhances the chance of men “abusing or attacking” women in discussions, how is this solved by a “safe space”?

          If i try to follow your reasoning, we should not want a “matriarchal” society, because that would enhance the chance that women will abuse or attack men in discussions (?). Or is the reasoning somehow different in the case of a “matriarchal” society?

          Regardless, i reason this “safe space” will probably not include those men that “abuse or attack” the women in this “safe space”. If this is correct, how is the possible problematic issue of men abusing or attacking women in an unscientific manner in scientific discussions solved?

  6. Quote from above: “Men in a patriarchal society will abuse and attack woman who engage in discussion in a certain way. Thus the need to build safe spaces for them to engage in discussion without being abused or attacked”

    This seems an incorrect statement, and/or a gross generalization.

    It could be that in certain forums, or at certain times, or in certain instances, a man could attack a woman in an unscientific manner, but this of course does not (need to) happen always (or merit the use of the word “will” in your quote).

    If, and when, that happens though it seems crucial for others to point to the possible unscientific, or even abusive, tactic or communication.

    Just like, others should point out when someone says, or writes, something that could be unscientific, untrue, or severely flawed.

    I reason, this will also help to make “safe spaces” less and less necessary, will help to keep everyone involved in the discussion (and not create “bubbles” or separate groups), and will help to keep scientific values and principles (also concerning debating) intact.

    • Jordan:

      I followed the link and I agree with Vazire on this. Another way to put it is like this: Speculation is great, but then please (a) label speculations as such, rather than labeling your published claims as discoveries and then downgrading them to speculations only after they’re questioned, and (b) value all speculations, not just those that happen to have “p less than 0.05” or “p less than 0.10” in some pile of noisy numbers, and not just those endorsed by insiders clubs such as PNAS or Perspectives on Psychological Science.

    • What i find worrying, and sad in some way, is that it (i feel) regularly only takes a few “big names” to start a certain discussion about a certain topic, and before you know it half of social science twitter is talking about it, writing their own personal blog posts about it, tweeting about it, etc.

      I am even participating in it myself in some way, by talking about it here (albeit on a more general level).

      My guess for the next 5 years is that a certain group of scientists will continue participating in all of this, and a certain group will (begin to) see what this is really doing to them (and perhaps even science in general) and stop (or severely limit) their “online presence”.

        • Quote from above: “I know what you mean, and that’s one reason I don’t tweet, and that’s one reason I blog with a six-month delay.”

          Yes, i am aware of those things. I think they could be a really smart way to go about things.

          I myself am currently trying to quit this whole science thing. I think it has been, and still is to a lesser extent, some sort of addiction for me. And i don’t even do “social media”, or facebook, etc. I don’t even have a phone!? I can only imagine how this could all be for scientists who have a phone…

          Anyway, concerning trying to quit my possible science-addiction: i intend for this blog to sort of be the scientific version of AA-meetings, or nicotine-patches. The last step before walking away from it all. Perhaps this blog could help me out in this way as well, as it did with several other things (which i really appreciate).

        • Anyway, concerning trying to quit my possible science-addiction: i intend for this blog to sort of be the scientific version of AA-meetings, or nicotine-patches. The last step before walking away from it all. Perhaps this blog could help me out in this way as well, as it did with several other things (which i really appreciate).

          What is going on here? There seems to be a story and you seem to want to tell it. So what’s up?

          Also, your diction strikes me as unique. Is there something up with that? Your use of hyphens also seems unique.

        • Quote from above: “What is going on here? There seems to be a story and you seem to want to tell it. So what’s up? Also, your diction strikes me as unique. Is there something up with that? Your use of hyphens also seems unique.”

          I just like this blog a lot, and (i think) the majortiy of the people participating on it. I view it as the university experience i never had, but wanted to have. I view it as what i think the rest of science should be like, and have found out is not.

          That’s part of what i mean with this blog being my “AA-meeting” or “nicotine-patch”: at this moment in time, it is the only “scientific” source i want to read, and the only place i want to participate in “scienctific” manners. Perhaps because it leaves me with some hope, and/or a sense of sanity.

          As for the use of hyphens, i use them a lot to emphasize words that i feel are important. I also use them to quote stuff.

        • Quote from above: “As for the use of hyphens, i use them a lot to emphasize words that i feel are important. I also use them to quote stuff.”

          O wait, i’m not so sure now what you meant with “hyphen”. My reply above related to what i believe is also called “quotation marks”. I use them a lot to emphasize words that i feel are important. I also use them to quote stuff.

          If you are refering to the use of the “-” -sign in words like “AA-meetings” and “nicotine-patches”, i use them because i though that is how those words are written.

          English is not my 1st language, and i don’t particularly necessarily care for the “proper” way to write things (to a certain extent). Perhaps the combination of those 2 things (and probably a few other things) could explain why you could think my writing is “unique”.

  7. I went all the way back to Dubner’s response and came away with a somewhat higher opinion of the Freakonomics blog. He made sure to quote his critics and respond to their specific charges (broken up into seven parts in your case), while your “meta” response sidestepped all that. Dubner doesn’t say anything about being hurt/intimidated by how nasty your criticism was (or call you “methodological terrorists”), just compelled to respond based on how inaccurate you & Blattman were. Beforehand you noted that critiques are worth reading as they may “contain worthwhile feedback, or point to an error in need of correction”, which I’d gather fits your conception of how the discourse works. In one instance he complains that you refer to them as neglecting to name the scholars who get it right when Oster got it wrong, even though his blog linked to articles citing those critiques. This is sort of the converse of Blattman’s complaint that they would write out a hat-tip to him but neglect to actually link. In either case someone could be dinged a little and it wouldn’t be that odd for them to respond that their blog post gave credit in one or another way, although the ideal would be to do both. That’s a matter of norms for blogging, and I think most of us expect that a blogger bringing attention to an academic paper (such as the examples you bring up in your “meta” response) has given it less thought than whoever wrote it. Dubner does seem interested in whether they got things wrong (as in when they repeated the inaccurate claims of Stetson Kennedy), but in that case his response is just to publish a correction rather than to take stock and ensure that doesn’t happen again (I’m not sure how easy that would be in the case of Kennedy, whereas you give reasons they should have been a priori skeptical of Oster). In other cases he doesn’t consider their work to actually be wrong as there’s no objectively “right” answer as to what assumptions to make when analyzing drunk walking vs driving or the acceptable Type I vs Type II error rate for detecting terrorists.

    • Tggp:

      I’m sure Dubner means well and is working hard to improve the world, but I don’t think he responded well to what Kaiser Fung and I wrote. For example, he labeled our writing as “weaselly” which adds nothing but pure negativity to the discourse. He also didn’t address the fact that Freakonomics repeatedly has pushed crappy research, including the claims about sex ratios, ESP, and Tiger Woods endorsements. He also personalizes our criticisms and attacks us, which misses the point, given that lots of people have criticized Freakonomics on the grounds that it is sloppy. Kaiser and I were not the first to make this criticism, not by a long shot. Our contribution was to discuss the issues from the perspective of statisticians rather than as economists or journalists, which is where other criticisms have come from. Finally, I wrote a meta response because by that time I was a bit tired of the whole thing. I’d explained the technical points enough times, and Kaiser and I wrote that article, to which Dubner responded rudely, so it seemed more relevant at that point to discuss the larger issue of how to have such discourse in the first place, rather than to repeat the discussions of what went wrong in all those examples we’d discussed many times.

  8. There is also safe space in data science, which hits right at home! A while back, I attended a meetup in which the speaker talked about deep learning models in economics. Not surprisingly, someone in the room – presumably a statistician – asked the O question. Most in the audience are academics or holders of graduate degrees. When I got home, there was an email sent to the entire group along the lines of: “in every one of these meetings, some idiot statistician asks about overfitting. I propose we ban this question from future meetings.”

    On the other end, I find that I like to make arguments while debating ideas in my head, play devil’s advocate, etc. but frequently, the other side assumes what I say is how I feel. So in effect, I’m demanding a safe space to put out ideas that I reserve the right to take back when it doesn’t work out. However, like other commentators, if the ideas are published in peer-reviewed journals, they ought to have been thought through!

    I haven’t read the Dubner response to Andrew and my article for a while but I do remember finding it unconstructive and not forwarding the debate.

  9. I think the discussion is partially confounded by some people assuming criticism/debate is always OK, regardless of form, while others (including me) requiring some basic civility in scientific discourse. The latter group is than being accused of wanting “safe spaces” and avoiding criticism, while all what is asked for is to be kind – especially when delivering criticism.

    My current understanding is that finding good arguments is the easy part of a debate. Getting the sides to actually listen to each other and cooperate on honest pursuit of the truth is the hard part. And a lot of online discourse is just too personal and nasty to make honest cooperation possible. I am all for “safe spaces” where we respect humanity and dignity of others and listen instead of shouting.

    Note: there are obviously caveats and some “civil disobedience” may be due against powerful actors that act in bad faith. But kindness should IMHO be the very default.

  10. The political theorist Jacob Levy makes some interestingly parallel remarks about safe spaces at
    http://bleedingheartlibertarians.com/2016/03/safe-spaces-academic-freedom-and-the-university-as-a-complex-association/

    I’m thinking in particular of this paragraph:

    “””
    one thing that’s notable about academic freedom as I’ve just labeled it is that it creates safe spaces. The people who are doing the work on an ongoing basis be they students, teachers, or researchers, don’t have to spend all day every day answering the challenge: “where are your experiments?”. That get’s boring and is unproductive. If I as a political scientist want to do some research if I want to make some intellectual progress, I have to not be constantly harangued by the biologists or the chemists saying: “It’s not a real science you know”. By the third time, there is nothing new to learn, it’s going to get in the way of our ability to do what we are doing. The same by the way is true of the metaphysicist in a philosophy department who says to the physicist: “Well you know that your assumptions about the nature of reality are really up for contestation.” It’s really hard to argue with and the physicist replies: “Get out of my way, I’m trying to get some work done!”
    “””

Leave a Reply to Kaiser Cancel reply

Your email address will not be published. Required fields are marked *