A Heideggerian pedagogy of disruption

Abstract The phenomenological tradition developed sophisticated techniques to draw attention to pre-theoretic or pre-reflective experience. This article examines how one of the most famous, Heidegger’s ‘broken tool’, might work in a pedagogical context. I contend that it can be highly effective there, fleshing out his vision of teaching as ‘letting learn’ with a distinctive educational method. At the same time, that context suggests fundamental changes to the standard reading of the ‘broken tool’, shifting the focus towards what I call ‘information tools’. My conclusions link to recent research stressing disruption in the learning process, and I close by contrasting my position with ‘bricolage’ theories and work by Applebaum and others.

practices that characterise the vast majority of our behaviour as we walk to work, chat with friends, eat lunch. 2 Heidegger's aim is to illuminate the meaningful structure of such everyday life.
To do so, he introduces a characteristically artisanal scenario: a craftworker, absorbed in their task, silently discards a hammer that is too light and takes a better one (BT:69,157). The case foregrounds several points: everyday action occurs fluidly, confidently and unreflectively within a meaningful context, tacitly structured by my goals and the equipment around me (BT:70;129).
At the core of this web is what Heidegger labels the 'for-the-sake-of-which': the agent's self-understanding, where that is analysed in terms of her aims, the skills she exercises in pursuing them and the shifting configuration of the world which is thus manifest to her (BT:87). Sally, for example, understands himself as a teacher insofar as she encounters and evaluates the everyday world in terms of that identity. Here is Heidegger's description of the phenomenology: Coming into the lecture-room, I see the lectern … What do 'I' see? Brown surfaces, at right angles to one another? No, I see something else. Is it a largish box with another smaller one set on top of it? Not at all. I see the lectern at which I am to speak. You [the students] see the lectern from which you are to be addressed and from which I have previously spoken to you … I see the lectern in one fell swoop, so to speak, and not in isolation, but as adjusted a bit too high for me. I see -and immediately so -a book lying upon it as annoying to me (a book, not a collection of layered pages with black marks strewn across them). (Heidegger, 1999, pp. 71-2) The result is a picture on which self and world go hand in hand: objects are encountered as means to specific ends, tasks as desirable or obligatory, all ordered, or in Heideggerian terms 'disclosed', by the agent's self-understanding (BT:192). 3 Many contemporary Heideggerians, drawing on Gibson, express this by talk of 'affordances and solicitations' (Gibson, 2014). For Sally, the pile of papers teetering on the desk 'affords' reading, whilst for the cleaner, it is simply an obstacle.
Heidegger claims that these structures underpin our everyday behaviourand yet are normally hidden from us. Unnoticed and neglected by traditional philosophy in favour of more theoretical questions, they are too close to see, just as I look 'through' but not 'at' my glasses. This is a point common to all phenomenologists: as Carman recently put it in discussing Merleau-Ponty, the target of phenomenological analysis is 'so basic and so familiar that we are normally unaware of it, so inconspicuous and so transparent to our ordinary … sense of ourselves as to be invisible' (Carman, 2008, p. 101). 4 The question therefore arises: how can these structures be made visible? The task, as §16 of Being and Time puts it, is to explain how this 'Worldly Character of the Environment Announces itself in Entities Within-the-world' (BT:72). And this brings me to the famous broken tool.
One key component of the everyday is the 'ready-to-hand', which Heidegger defines in terms of equipment: 'the mode of being which equipment possesses … we call 'ready-to-hand'' (BT:69). As ever, Heidegger's taxonomy is complex, but for the moment we can focus on the main example he gives, that of physical tools (BT:70-1). Heidegger's basic idea is this: when a tool suddenly fails, that failure makes visible the network of tasks and goals in which it was tacitly embedded.
For example, it is only when the internet suddenly drops that we realize how interwoven it has become with every aspect of daily life. Here is the claim in his own terminology: But when an assignment has been disturbed -when something is unusable for some purpose -then the assignment becomes explicit … The context of equipment is lit up, not as something never seen before, but as a totality constantly sighted beforehand in circumspection. With this totality … the world announces itself. (BT:74) The same, he suggests, occurs when a tool is missing: it is when I can't find my phone that its importance to every aspect of my life becomes brutally clear.
Similarly, when something ready-to-hand is found missing, though its everyday presence [Zugegensein] has been so obvious that we have never taken any notice of it, this makes a break in those referential contexts … [We] now see for the first time what the missing article was ready-to-hand with, and what it was ready-to-hand for. (BT:75) In short, disruption to the everyday is a vital epistemological device: it is this disruption which lets us explicitly see the tacit web of tasks, goals and equipment that underpins our lives. 5 Is  (Golob, 2014, p. 164). The mathematician's favourite lemma is just as much a tool of the trade, linked to her self-conception and tasks, as the carpenter's lathe. Let us see what happens if we consider 'information tools': the cognitive devices which we rely on for everyday, rapid processing of information.
One classic example is the stereotype, i.e. a representation of some generic class in terms of prototypical exemplars. Stereotypes can, of course, be dangerous insofar as they project the qualities of individuals onto a group. But my definition is neutral: what is important is that stereotypes offer speedy, low energy information processing. Here is Kahneman making the point in terms of his famous 'System 1' and 'System 2' distinction, where System 1 is designed for rapid, non-deliberative decision making.
Stereotyping is a bad word in our culture, but in my usage, it is neutral. One of the basic characteristics of System 1 is that it represents categories as norms and prototypical exemplars. This is how we think of horses, refrigerators, and New York police officers; we hold in memory a representation of one or more 'normal' members of each of these categories. When the categories are social, these representations are called stereotypes. Some stereotypes are perniciously wrong, and hostile stereotyping can have dreadful consequences, but the psychological facts cannot be avoided: stereotypes, both correct and false, are how we think of categories. (Kahneman, 2014, p. 165) Kahneman is merely the latest proponent of a tradition within psychology on which stereotypes are treated as 'energy-saving devices'. As Macrae, Milne and Bodenhausen put it: Stereotypes, accordingly, serve to simplify perception, judgment, and action. As energy-saving devices, they spare perceivers the ordeal of responding to an almost incomprehensibly complex social world. (Macrae et al., 1994, p. 37) In other words, stereotypes are the 'go to' tools of much of our social processing: Gilbert and Hixon called them the 'tools that jump out' of our cognitive toolbox (Gilbert & Hixon, 1991, p. 510). 6 I now want to rethink Heidegger's argument with stereotypes in place of hammers. Suppose we are using this particular information tool and that it somehow 'breaks'. Does this make sense and if so what would it amount to?
Well, suppose I am happily and unthinkingly using the relevant tool. Then suddenly new evidence is presented which reveals how badly the stereotype has misled me. The shortcomings of the tool are thus placed center stage. In extreme cases, its distorting effects could be shown to be so high that, despite its low resource costs, I resolve to abandon it. How do such disruptions or breakdowns differ from those of physical tools? The basic answer is that, all else being equal, the disruption will be greater. This is not because stereotypes frequently deal with delicate social situations whilst the broken hammer only impairs carpentry: there are obviously physical tools whose failure is a matter of life and death. Rather, it is because of four structural features of information tools.
First, the failure of an information tool affects both past and future cases. If the hammer head suddenly comes off in my hand, this does not call into question all my previous acts of hammering: their success or failure was perfectly easy to ascertain at the time. But I might now suddenly wonder if previous interactions were, unbeknownst to me, misled by the 'broken' stereotype. Indeed, there is a sense in which those previous interactions are never 'over and done with': insofar as I remember them, they feed into my current narrative and assumptions and are still being processed by the same stereotype tool. Again, this does not apply to the hammer: the fact the head now comes off gives no reason to think that long past acts of hammering must be revisited.
Second, the failure of a stereotype tool will often call into question not just past and future uses of that same tool, but a whole range of other inferentially linked tools, concepts, and judgements. The relationship between low resource/high speed cognitive tools and more explicit forms of judgment is massively complex, but given there is a link, the failure of a stereotype tool will have inferential implications. If Joseph's stereotype of police as friendly and protective suddenly 'breaks' in watching the George Floyd footage, that inevitably raises questions for him about their social role, about other instances in which he has sided with police accounts and so on. Due to the inferential links between concepts, when an information tool fails, the disruptive effect will, all else being equal, be greater: when my hammer breaks, in contrast, my saw or my screwdriver continue to function without impairment.
Third, the failure of an information tool will typically lead to larger disruption in my interpersonal relationships. Of course, there are physical tools whose failure would have vast consequences. But the genesis of information tools means that, all other things being equal, their failure is particularly disruptive to interpersonal relations. This is because I receive many such tools from the community who raised me: their failure thus has the potential to call into question both the donors ('was Dad really a bigot?') and the many other tools I got from them ('what else did he teach me that might be wrong?'). In contrast, physical tools are less likely to be acquired from a single source to whom I have intense attachments. Of course, it might be that every physical tool I own is a family heirloom or that the head coming off a hammer makes me regard the saw as suspectmy point is that is very much an outlier case.
Fourth, the failure of an information tool such as a stereotype will typically lead to greater disruption in my own self-conception. In Heidegger's terms, my 'for-the-sake-of-which' is standardly unaltered by the failure of the hammer. Information tools, in contrast, are more closely bound with my own self-conception: one crude reason is that they lack the externality of material objects and so are easier to identify as part of 'me'. Another is that their failure often exposes a range of moral or ethical failings: it isn't simply that I can't hammer a nail, but that I have systematically treated others wrongly, perhaps despite good intentions. This will bring with it greater levels of personal disruption, for example self-reproach.
I have introduced information tools as an adaption of Heidegger's argument. At a technical level, it marks a significant change. The ready-to-hand for Heidegger denotes entities within the world: cars, hammers, saws, etc. In contrast, information tool are, in his terms, part of Dasein, part of the disclosive structure of the subject rather than an entity to be disclosed. Nevertheless, the spirit of his account is retained. The key Heideggerian claim is that tool failure illuminates the 'worldhood', the web of instrumental and self-interpretative relations sketched in Section 1. When a stereotype fails, there is potential for exactly the same illumination. A sudden moment of disruption illuminates the tacit processing device which I have been using and the way in which it fits into my own self-conception: white viewers whose stereotype of the friendly police officer 'breaks' as they watch the Floyd footage simultaneously find themselves having to ask why their own interactions with the police had proceeded smoothly.
My primary aim here is not exegetical: I am not centrally concerned with whether Heidegger would personally accept this reading. Rather, my claim is that the 'broken tool' analysis can be extended to a class of tools with far greater social importance than the original hammerand that doing so brings a number of interesting results.

Applying the argument in a pedagogical context
The preceding section reworked Heidegger's argument in terms of information, rather than merely physical, tools. The task now is to apply this to the pedagogical case.
Imagine a teaching practice in which scenarios, either fictional or real, are used to activate and then disrupt stereotypes in the way sketched: i.e. to first trigger their unthinking, fluid deployment as an information tool and only then suddenly to cause their failure, by exposing their misleading effects. This has two attractions. First, as in Heidegger's original account, the moment of disruption makes visible the world and the tool's place in it: it becomes clear how much it was being relied on by its users. Second, unlike Heidegger's original account, the aim is not simply descriptive: it is not simply to highlight the world and the tool for further philosophical description. Instead, that process involves improving the tools we use, as the teacher models a series of scenarios designed to expose the ways in which they mislead.
Of course, there will have to be a separate debate as to which stereotypes are the most misleading and my account is completely neutral on that. Its point is to identify a device, the moment of sudden disruption as a tool 'breaks', that can first illuminate and then improve our information capacities and our awareness of the 'world' structure in which they operate.
To flesh out the story, let's turn back to Heidegger's original 'physical tool' version. For all its acuity, several aspects are phenomenologically problematic.
First, as Blattner observes, Heidegger makes it sound as if we gawp helplessly at the broken hammer as the disruption: [S]evers one's insertion into the means-ends logic of the everyday world and leaves one staring at a pile of stuff bereft of its practical involvements. (Blattner, 1995, p. 324) Of course, this is not the case: we normally engage in 'work-arounds', adjusting to find another means to our end. When the car doesn't start, you begin running for the bus: you neither simply stare at it nor start contemplating transport in the modern world. Dreyfus has responded that such work-arounds are only available in cases of 'minor' disruption: in contrast, it is 'total breakdown' which makes 'worldhood' visible (Dreyfus, 1991, p. 81). But that is not right either: people typically react to catastrophic failure by succumbing to shock, or desperately scrabbling for stability, rather than launching into philosophical analysis.
Of course, one could imagine someone with a particularly reflective cast of mind doing the latter. But, and this brings me to the second point, in such cases the whole breakdown story seems unnecessary: why couldn't such an individual come to think about these issues by reading philosophical texts or by looking around with an abstract curiosity? Heidegger's tool story seems neither sufficient nor necessary for the epistemic pay off he promised.
Third, Heidegger focuses on tools which we regard as disposable: faced with their breakdown, we would 'like to shove them out of the way' (BT:74). But agents are frequently highly attached to the equipment they depend on: it matters that they drive their car or use grandpa's pen, and they respond to disruption by 'caring for' the tool, trying to patch it up, rather than by shoving it aside. Again, we might wonder if Heidegger's story is really true to our experience.
Turning back to the information tool model, we can see how it handles all three points. First, exactly as Blattner predicted, our typical response to information tool failure is to seek a 'work-around', a way of preserving the tool by dismissing the disruptive event as anomalous ('just a few bad apples', 'no true Scotsman' etc.). This possibility is ever-present and I am certainly not claiming that any one-off experience will easily rid agents of long-held assumptions.
However, the disruptive breakdown experience provides a distinctive window of illumination, bringing the tool's role into focus, and of possible development, assessing its costs. Using this window will require careful pedagogy. For example, one task for the teacher will be to break down rationalisations as they occur: to borrow a phrase from Herbart, this requires the 'pedagogical tact' to avoid triggering a further 'digging in'. 7 Following Heidegger himself, we can take this idea of a 'window of illumination' further. 8 Heidegger frequently laments education's focus on mere 'data', rather than on a deeper 'unhiddenness' (Heidegger, 2002b, p. 53). 'Information tools' constitute such unhiddenness because they determine how we process the world: they frame our view of the data, rather than being simply another piece of it. At the same time, however, the window afforded by breakdown is easily ignored or slammed shut: unlike Aristotelian accounts which often stress linear moral development, breakdown opens the possibility of insight but in itself, especially unsupported by 'pedagogical tact', it remains only a possibility.
Second, I noted that you might doubt whether Heidegger's method was necessary: why couldn't I simply read a philosophical account of world or think a bit about the role of hammers? Why is breakdown so important? We can now give at least one reason: unlike the craftsman's hammer, stereotypes are tools which their users frequently deny possessing. It is therefore hard to evidence how widespread they are and the myriad roles they play in structuring our experience simply by reflection. The sequence of unthinking use followed by breakdown and disruption is valuable as it creates both an illustration of those roles and a subsequent moment of disorientation that allows for a possible opening, the window I spoke of above. My 'information tool' model is thus better able than Heidegger's original to explain why breakdown is distinctively important.
Third, I noted that Heidegger's original failed to capture the phenomenon of 'tool care': the lengths to which individuals go to repair damaged equipment, rather than simply shoving it aside. My information tools story handles this very naturally: as noted in Section 2, we feel that information tools are more intimately part of us than any physical counterpart. To simply 'shove aside' a damaged information tool would be to repudiate a whole strand of our past life, moral and epistemic. The result is tool care, of which rationalization is one of the most common and potentially dangerous forms.
My aim has been to show how we might think about disruption in terms of information tool breakdown, adapting one of the most influential epistemic devices within phenomenology to a pedagogical context. This has simultaneously provided a new standpoint on Heidegger's original argument. Throughout, I have been deliberately neutral on the details of how such information tool 'breakdowns' might be produced. This is because I do not want to commit myself to the viability or otherwise of specific devices such as Harvard's popular Implicit Association Test. Similarly, my aim is not to advance any particular curriculum programme. For example, I am not making any claim as to what might damage or disrupt a given tool: which images or texts or scenarios can do the trick will be highly culture-dependent, and educators will need to make a judgement on a case by case basis. My goal is to sketch a broad pedagogical model: teachers can then hone the specifics based on the contours of their own classroom situation.

Contrasting the proposal with existing pedagogical approaches
Perhaps the best way to bring out the pedagogical implications of this proposal is to contrast it with some prominent alternatives. In particular, I highlight the difference from other approaches that stress disruption as part of the learning process.
First, the proposal differs from another classic phenomenological case: a sudden illness which makes me notice overlooked features of the environment. My sore leg, for example, makes salient just how many stairs the station has. As with a failed information tool, such limitations will often prompt feelings of vulnerability as a previously reliable capacity fails. But the focus in the illness case is on the surrounding environment: the epistemic insight is not that my leg is sore, which I knew, but that the station is inhospitable to all but the able-bodied. In contrast, my model focuses on the information tool itself: the novel epistemic payoff concerns my capacities and their shortcomings. The two approaches are thus distinct but complimentary.
Second, consider bricolage theories. In an educational context, 'bricolage' is sometimes used very broadly to mean any approach drawing on several methods (for example, Guatam & Lowery, 2017). But in one central version it refers to a more specific approach: one in which a given tool is held fixed whilst the surrounding instrumental context is varied. Consider Sohn-Rethel's famous case of a 'young man on the street in Capri who transformed a motorcycle engine into a device that makes whipped cream' (Sohn-Rethel, 2009, p. 111). Here the tool, the engine, functions perfectly but the context has changed. Heidegger's original example runs in the opposite direction: the carpenter is still trying to build a table, but one piece of equipment has failed. My approach starts like Heidegger's: the habitual tool a student uses is disrupted or damaged. But, unlike in Heidegger, this does not simply illuminate the pre-existing instrumental totality. Rather it creates a window for change, as the tool's role and its shortcomings become salient and open to critique. Of course, Heidegger himself has plenty of critical things to say elsewhere but the focus in Being and Time §16 is fundamentally descriptive: to illuminate the world. 9 On my model, in contrast, the focus is simultaneously the tool, its place in our world and possible alternatives to it.
Third, I want to compare my approach with recent work in what is often called the 'pedagogy of discomfort'. This stresses the necessity of student discomfort as part of the learning process in the context of a social justice agenda (prominent examples include Applebaum, 2017;Berlak, 2004;Boler, 1999;Kumashiro, 2002).
There are certainly points of contact: as noted, the failure of information tools will typically produce a defensive and uncomfortable reaction, and it will require careful pedagogical engagement or 'tact' to avoid a slide into rationalization. However, my approach is phenomenological in a sense which such theorists would not endorse. This is because, like Heidegger's original model, it operates from out of the students pre-existing world and equipment: the process requires them to be using a familiar tool and then to be faced with a breakdown in it. They must start off cognitively at home, as absorbed and comfortable as Heidegger's carpenter in the workshop. In contrast, accounts such as Applebaum's begin from a teacher-introduced theory and the problem of discomfort arises when some students react to this as an alien and hostile attack. Here is Applebaum's original description: Last year my colleague invited me to visit her class and address the topic of 'Discourse, Truth, and White Strategies of Denial'. After my presentation, a lively conversation ensued around white denials of racism and complicity that was led primarily by the students of color in the class … Noticing that the white students in the class were silent, I pressed them to engage with what the students of color were saying. (Applebaum, 2017, pp. 862-3) This can be a valuable approach, but it is a different one from mine. On my account, the discomforting and illuminating effect comes from the sudden transition from a comfortable 'everyday' experience to the feeling of breakdown. As a phenomenological approach, my account begins not from a piece of theory, but out of the lived world of the student. The result is a very different classroom structure. The starting point is not the teacher's presentation of a theory, nor is it even the teacher eliciting narrative or quasi-theoretic testimony from disadvantaged or minority students. Instead, it is the use of student-centered scenarios which lull individuals into deploying the range of stereotypes in question.
Of course, this raises social questions that I cannot answer here, for example about the rights of individual students. My aim is a modest one: to introduce a particular technique and way of thinking about moral illumination.
The contrast with social justice approaches is worth pressing since it clarifies the distinctively Heideggerian nature of the proposal. Heidegger argued that teaching should consists in 'letting learn' (Heidegger, 2002a, pp. 17-18). What I have offered is an account of a learning and classroom process that meets that ideal: one which operates from out of the student's everyday information tools and which advances as the student encounters the failure of those tools. In contrast, Heidegger saw dominant models of education as 'theory-led' and 'hierarchical', where the teacher's role is to dispense abstract knowledge which the pupils lack, using concepts unavailable in their pre-theoretic experience (Heidegger, 2002b, p. 39). Approaches such as Applebaum's fall into that category: one on which the teacher, armed with a complex theoretical apparatus guaranteeing epistemic superiority, determines the discussion. There are, undoubtedly, attractions to both techniques. The Heideggerian approach roots the discussion in the students' prior conceptual world and works out from that: if you believe this conceptual world to be irredeemably contaminated, such a starting point, characteristic of hermeneutic philosophies, will be problematic. But my aim is not to establish the superiority of one method over the other; indeed, I am sceptical of whether blanket claims for superiority make sense. Rather, it is to make clear the differing commitments of the two models.
Another way to put the point is that my account does not privilege disruption because of links to social justice, but because of its vision of learning: one in which the disruptive breakdown of everyday information tools by their own users drives the learning process. I focused on correcting social stereotypes not because they are essential to the modelthat would be surprising given Heidegger's own politicsbut because they provide one striking and important illustration of how we might use it now. 10 This is not to say, however, that it can work for any subject matter and in domains where it is unattractive the reasons mirror the method's substantive commitments. For example, thermodynamics, as least as standardly taught, involves the grasp of abstract principles governing a 'scientific image' famously different from the 'manifest' one of the familiar, everyday world. As Heidegger repeatedly observed, physics' conception of space is not one in which we are standardly at homeand thus not one which lends itself to a learning that works out from our everyday lives (for example, Heidegger, 2000b, p. 157). Instead, it is apt for hierarchical, theory-led teaching: the physics teacher may give illustrations of how stated principle impacts on everyday life, but no standard thermodynamics course aims for the students to derive those principles by working their way out from everyday life. 11 So my approach, precisely because of its substantive commitments, will not suit all areas.
Any process of ethical education is likely to be a long one, requiring a wide range of resources. I agree entirely with Zdenek and Schochor that 'teachers must do more than simply provide opportunities for moral dissonance to foster moral development in students' (Zdenek & Schochor, 2007, p. 520). My claim is not that the broken tool model is any kind of magic bullet, but that its distinctive blend of phenomeology and critique makes it another useful weapon in the conceptual armoury and one that can help us rethink Heidegger's arguments and his contribution to education.
Notes present-at-hand' (BT:75; BT:73). The problem is that 'presence-at-hand' is enormously ambiguous: Golob 2014 distinguishes three meanings in Heidegger's work, whilst McManus (2012) catalogues more than thirty-seven uses (Golob, 2014;McManus, 2012, pp. 53-7). For example, one way to read him is to equate 'presence-athand' with the objects of the natural science: a breakdown in our everyday experience would supposedly support both an awareness of its structures and a transition to the kind of decontextualized analysis of entities found in mathematical physics (Dreyfus 1991, p. 81 is the classic Anglophone example). Assessing such arguments, however, would take us too far afield: I therefore bracket the question of presence-at-hand and focus on Heidegger's use of disruption as an epistemic device. 6. There is a vast debate surrounding the attendant philosophy of mind and psychology, and nothing in what follows rests on any particular theory of stereotypes. Indeed, if you prefer you can replace that term with some other rapid processing, minimal resource information device: from Kantian schemata to Kahneman's 'automatic system 1' there are plenty of options. What matters is simply the idea of a fast processing, low cognitive resource information tool: since stereotypes are relatively familiar, I use those as my example. 7. For an extremely helpful discussion of Herbart's position which meshes with the arguments here see English, 2013. 8. My thanks to an anonymous referee for pressing me on this. 9. In telling Heidegger's own 'critical story', anxiety, a disruption in which the entire world suddenly seems insignificant (BT:186), plays a crucial role. I have not discussed anxiety because Heidegger's vision of it is tied to his larger picture of 'Dasein', and it requires a number of assumptions from that picture. As Withy nicely puts it, 'Heidegger takes human existence to be uncanny' (Withy, 2015, p. 2). Instead, I have focused on a technique that can be useful without a commitment to the rest of the evidently controversial Heideggerian story. 10. I am indebted to an anonymous referee for helping me work through this aspect of the argument. Another example illustrate further complexities stemming from the way phenomenology interdefines 'everyday' and 'theoretical'.
Jo and Frank are obsessively dedicated scientists for whom the concepts of advanced physics do indeed form part of their 'everyday world'. Faced with a radically unexpected result, Jo concludes that an unquestioned basic category is in fact faulty. Frank, due to strong identification with the particular school that pioneered that category, insist that the data must be erroneous.
This Kuhnian situation is amenable to my analysis: an information tool has broken, opening both a new epistemic window and a space for defensive rationalisation. As this shows, from a phenomenological perspective, what counts as 'everyday' and 'theoretical' depends on one's lifeworld (Heidegger 1983, p. 243): from their fluid confidence in the tools to their personal investment in the theories, Jo and Frank, as experts, fit my non-hierarchical model of learning in a way that classroom physics lessons do not. Note incidentally, that there is no link to social justice issues: the role of disruption in my account is again quite independent of the example of social stereotypes. 11. It is possible such a course could be developedfrom a Heideggerian point of view, that would be a metaphysically significant innovation in the pedagogy of physics.