The Challenge to Educational Reforms during a Global Emergency: The Case of Progressive Science Education

• This article argues that what is most at risk in schooling during a global pandemic, or other similar broad challenges to normal functioning, are those elements that might be considered the less traditional and so the most progressive. After setting out some general background common to the challenge faced by schools and school teachers, this argument is exemplified through the case of school science education. Two particular aspects are considered: one related to pedagogy (responding to learners’ alternative conceptions or ‘misconceptions’) and one related to curriculum (teaching about the nature of science). These are considered ‘progressive’ features in the sense that they have widely been championed as ways of improving and reforming science education across a wide range of national contexts but can be understood to have faced resistance both in the sense of being opposed by ‘reactionary’ stakeholders and in terms of the level of support for teacher adoption. It is argued that at a time when the education system is placed under extreme stress, such progressive elements are at particular risk as teachers and administrators may view them as ‘extras’ rather than ‘core’ features of practice and/or as reflecting more ‘difficult’ educational objectives that may need to be de-prioritised (and so neglected) for the time being. In that sense, they are fragile aspects of practice that lack the resilience of more established, and thus robust, features. It is concluded that where progressive elements are especially valued, they need to become sufficiently embedded in custom and practice to no longer be viewed as luxuries but rather to be recognised as core elements of good teaching to be protected and maintained during a period of emergency.


Teaching relies on a social contract between teacher and learners
Even in an ideal context, in which a teacher and all her class are well connected, there are significant challenges to school teaching, both in primary (elementary) and secondary (high school) contexts. As most people today are encultured into societies with school systems, it is easy to overlook how schooling is far from a natural system of education. Specifically, humans evolved to be capable learners within certain social contexts -usually small groups whose members have graduated and progressing levels of expertise (Lave & Wenger, 1991). We still find something like this in the postgraduate education of scientists (Kuhn, 1970). The novice joins a specialist laboratory or research group as a new research student alongside other group members who are already further in their programme of study (established research students, post-docs who have graduated beyond that stage; and university lecturers and professors with considerable experience and expertise).
Such a context allows prolonged engagement with specific areas of learning, a high level of commitment to, and ownership of, a personal project, individualised learning paced for the particular learner, and opportunities to learn specific skills, techniques or ideas on a 'just-in-time' basis. By contrast, school (and much undergraduate) education is based on a model of one expert teaching many novices in short blocks of scheduled time. Of course, this is more efficient for mass education in logistic and economic terms (and the usually tacit childminding function of schooling has become explicit in public discourse during the pandemic), but means that students are often not learning something they are especially interested in, and to a large degree all those in a specific class have to progress through the curriculum together despite individual differences. In many ways, the successes of teachers in so often managing, motivating, and supporting student learning in school classrooms and lectures halls should be seen as an incredible achievement -relying on the strong interpersonal skills of teachers as much as their knowledge of the curriculum. The ability of a teacher to engage a diverse group of learners in a topic in which most have little intrinsic interest (as is often the case in school teaching) is something that can too easily be taken for granted -but, as new teachers often discover, is far from automatic.
What keeps students in the classroom, and hopefully paying attention, is certainly sometimes intrinsic interest in lessons, and may sometimes, in part, be a threat of some form of formal chastisement, but, often, is largely a kind of social contract between student and teacher. Teachers who are judged to be respectful to, and interested in, their students, and seem to care about them as individual people, and who clearly make an effort to give interesting and informed lessons, are usually rewarded with the default of most students not being disruptive, and, further, acquiescing in reasonable requests to undertake specific activities and to moderate the natural human tendency to chatter at will.
Once the 'social contract' of the classroom is negotiated (whether explicitly or tacitly) and once a good working relationship is established, both 'sides' will earn some credit to be forgiven some occasional lapses without this being seen as a threat to the established norms. A student will be forgiven the uncharacteristic slip -a misjudged joke, a yawn, a few minutes daydreaming -just as the students in a class will understand and forgive the usually fair, reasonable, and conscientious teacher who is very occasionally ill-tempered or does not seem to have prepared well for a particular lesson, or who tries something new that does not seem to be working. A teacher who, obviously, usually makes an effort to engage students in sequences of interactive activities can occasionally -on account perhaps, for example, of a headache or sore throat -persuade students to spend a lesson in quiet reading and note-taking that would otherwise be objected to as 'boring' . There is an (at least implicit) agreement: 'We may not get as much learning done today as usual, but we will have an orderly and peaceful classroom where I will tolerate some quiet chatter, and you will at least engage to some extent with the task I have set' .
Students do not generally put down their work and walk out of the classroom mid-lesson, or ignore the set task and engage in some unrelated activity for extended periods, even when they might be tempted, as this would be an overt contravention of the social contract and the teacher-student relationship on which it is founded. This restraint is, however, in part maintained by the nature of the setting. The teacher can normally see the whole class. Moreover, when the teacher is busily engaged with an individual or group of students, the classroom has something of the nature of the panopticon (Foucault, 1991(Foucault, /1977 in that the activity of the students is visible to their peers, and students will often join in the processes of monitoring and regulating the classroom (e.g., through announcements along the lines of 'Miss, Jenny is looking at her phone' or 'Sir, Tommy has put his books away, and there's still ten minutes left').

Changing the mode of teaching
Working 'online' is a very different proposition. When students are highly motivated to learn and make the best of the activity, for example, adult students who have enrolled themselves on professional development courses or postgraduate programmes, the teacher does not need to be so concerned about maintaining engagement. However, in a school teaching context, it is not as easy to monitor a class of 30 adolescents, each working on a device at distance, as it is when they are in the room with the teacher when eye contact can be made with any student in a moment. It is not so easy to notice someone who has absented themselves from the lesson or to see what the face apparently looking into the webcam is paying attention to on the screen. A child who leaves the computer and exits the room may do so covertly, without obviously breaking the usual contract. Leaving all microphones on at once is a recipe for noisy distractions -but muting microphones negates spontaneity of the usual classroom dialogue and a key mode for monitoring student activity.
Moreover, teaching online is unlikely to mean just doing the same lesson via computer. Many activities do not unproblematically transfer to home-based learning. Practical work in the sciences is an obvious case. Artefacts and models that may usually be manipulated cannot be engaged with as directly. In addition, key resources usually available in the classroom may not be available online.
That said, there are likely relevant alternative resources online that could be accessed. After all, the Internet gives access to the World Wide Web, offering a virtually unlimited range of resources. For most courses, it would be possible to find excellent, suitable resources online. When planning an online course, the identification and evaluation of resources would be a key task. However, that is not possible when suddenly being told that a course normally held in school or college is now to be interrupted and continued virtually. The sheer volume of Internet-accessible resources is matched by a diverse range in quality, and indeed a considerable level of misinformation. The curation of reliable, curriculum-matched, and correctly pitched resources is a critical task in planning teaching. Regardless, then, of any question of whether some material can, in principle, be taught online as well as in person; there is the issue of the time commitment for advanced planning of a coherent, well organised, and well-resourced course (Taber, 2018a): something that clearly can not happen when schools are summarily closed, and the mode of teaching switches, with virtually no warning, overnight. This challenge of switching modes for whole classes is exacerbated when working with classes split between those attending the school (and probably reorganised into novel collectives) and their classmates requiring teaching at distance.

Teachers develop expertise through specific teaching experiences
Teaching is honed over time. A strong understanding of subject matter is clearly important for effective teaching -as is a good appreciation of general principles of pedagogy and knowledge of the specific curriculum requirements set out as target learning for a particular course. Teachers not only need the pedagogic content knowledge (PCK) relating to common learning difficulties and teaching approaches in a topic (Kind, 2009) but arguably also develop a specialised form of their own subject knowledge through experience of teaching it to learners at a particular level.
So, for example, we might consider that academic chemists who research different areas (e.g., synthetic routes of natural products, as opposed to light-catalysed reactions or electrochemistry) each develop a particular form of subject knowledge which, although it may encompass the whole discipline, has particular depth, detail, nuance, and density of associations, focused on the area of specialist study. By comparison, the school chemistry teacher may seem a generalist but also develops specialised subject knowledge that is especially rich in relation to how the subject matter is processed in preparing and carrying out teaching. In relating subject knowledge to PCK (e.g., common misconceptions, useful metaphors and analogies, suitable simplified teaching models), the teacher also develops a particularly rich subject knowledge that is, in its own way, a form of specialism (Taber, 2020).
Knowing the subject and knowing how to teach are starting points but do not automatically lead to effective teaching. The teacher receives feedback through the practice of teaching: refining ideas about what works well, why a supposedly sensible sequence needs to be modified, how much longer a particular activity needs with a certain type of class, what level of understanding is reasonable to expect after first introducing a new concept, etcetera. Substantially changing the way in which teaching takes place acts as a kind of reset.
Just as when a new curriculum is introduced, or an innovative teaching approach adopted, a shift to a new mode of teaching changes the process: perhaps, a concept that had previously been readily explained suddenly becomes more opaque to learners, perhaps an activity that normally takes 20 minutes now only needs 15, perhaps paired or small group activity that usually works well at some point would be better substituted by something different. However, these are empirical questions that can only be addressed and indeed may only arise as teaching proceeds. Unfortunately, this is often ignored in experimental studies of teaching innovations. Instead, it is common to see well-established practice used as a comparison condition against some novel pedagogy, curriculum, or teaching resource that study participants are using in their teaching for the first time (Taber, 2019).
The global pandemic of 2020-21, then, meant that teachers not only shared in the common complications of the pandemic (risks to health, restrictions on travel and socialising, worries about at-risk relatives and friends) but also faced specific additional challenges in their professional work: including sudden shifts to less familiar modes of working, and the need to reorganise their lessons and courses without the time for advanced planning that is normally expected when making any substantive change to professional practice. In some cases, teachers may have been expected to simultaneously continue with planned teaching to reduced classes whilst also trying to offer the same curriculum to other students now working away from school. Thus, no matter how well-intentioned, committed, and hard-working teachers may be, the Covid-19 pandemic introduced challenges that will have impacted the quality of teaching and learning. Inevitably, when faced with such increased demands and new challenges, teachers will need to prioritise and adopt coping strategies. One colleague told me that a muchheard phrase in conversations between teachers was 'it is what it is' . Inevitably, some things that were previously recognised as important, desirable, and/or good practice will be casualties of the emergency.
This article explores what seems a reasonable conjecture: in an emergency situation (such as being suddenly required to 'deliver' the curriculum in novel and unfamiliar ways), what will be sacrificed to 'make do' will be those things seen as desirable but difficult. These are likely to comprise those elements of teaching considered 'reform' practices. What is understood here as 'reform' is that which is still widely seen as novel and challenging and so often perceived as 'difficult' and perhaps even as luxury. What is necessary (for teachers) is to teach the curriculum. What is by contrast seen as desirable is to incorporate those aspects of good practice that are still yet to be fully consolidated into 'custom and practice' and are still conceptualised as reforms. Another term that might be used instead of 'reform' might be 'progressive' . It is suggested that those features of a teacher's work that are still perceived as reforms or progressive are most likely to be less robust and less resilient in response to stressors. The scenario offered in this essay may be considered to present hypotheses that can subsequently be tested in research on the impact of the pandemic on education in various contexts.

Progressive science education
The term 'progressive' implies going beyond what is currently taken as standard fare or the norm. Formal education -such as schooling -is a social phenomenon depending upon cultural institutions. What is introduced as a reform and seen as progressive in one cultural (e.g., national or institutional) context may be viewed as unexceptional or conversely radical elsewhere. Indeed, in terms of educational reform, it is likely that there is a common pattern of a proposal being initially seen as radical (as 'left-field') before it is later adopted as a reform and considered progressive, and then later still becomes custom and practice (see Figure 1).

Figure 1
Initially fragile features of practice become more robust over time

Educational norms shift
For example, consider how curricula have shifted over time and still vary somewhat in different parts of the world. The medieval university curriculum was at one time dominated by the common study of the trivium (grammar, logic, rhetoric) and then the quadrivium (arithmetic, astronomy, music, and geometry), whereas today the norm is that undergraduates specialise, and from a much wider range of subjects such as chemistry, art history, sociology, civil engineering, and so forth. Moreover, whether it is appropriate to have university degree courses in subjects such as media studies, sports science or, indeed, education has at various times been the matter of debate (and subjects accepted in some universities or countries would not be in others). In the English system, an undergraduate would often focus on one discipline with a modest complement of subsidiary subjects (usually from fairly cognate disciplines). However, U.S. undergraduate courses often have a 'liberal studies' aspect such that a student may be required to study some science even if they are specialising in the humanities (Bourke et al., 2009). Chinese undergraduates are expected to study some aspects of a common curriculum such as mathematics, English, and state ideology (Zhang, 2012).
At one time, post-elementary school education in some countries took place in institutions known as 'grammar schools' -a term descriptive of their main focus, Latin grammar. No doubt, the addition of Greek would have initially been seen as a radical reform. The introduction of subjects such as the natural sciences into mainstream schools was also initially a progressive notion, which has become so taken for granted that any suggestion today that schools should not teach science would seem bizarre (and, now indeed, radical). Again, such a change was not uniform across the globe; for example, when the teaching of natural science was still seen as a novelty in the English school system, an official government report (Schools Inquiry Commission, 1868) not only pointed to where the innovation was being adopted around the country, but also to how the (more progressive, in this sense) French, German, and Swiss school systems were already embedding this curriculum reform.

Fragile features of school science education
What is considered progressive not only changes over time but is also relative to local norms. In this article, I will identify and discuss two aspects of science education that I will conjecture can be widely considered progressive. That is, these two features represent aspects of science education which a) have been much discussed and championed in the literature, b) have been incorporated into educational reforms in a range of national contexts (although not yet globally fully adopted), but c) are still recent or current enough reforms in many contexts as not yet to be sufficiently consolidated into custom and practice to be robust enough to avoid disruption at a time of substantial challenges to the system. They might be considered progressive features that are still 'fragile' (see Figure 1), meaning those that lack 'resilience' .
Space does not here allow an account of how these features have been adopted to various degrees in different contexts. Discussion of how these progressive elements have been nominally formally adopted in the English curriculum context, yet in a way too superficial to support teachers in deep engagement, can be found in other articles (Taber, 2010(Taber, , 2018b. One of these fragile progressive features relates primarily to curriculum and the other to pedagogy. I will describe each of these features, with some background on the arguments for their adoption in school science, and discuss why they might be considered fragile and so vulnerable when the school system is highly stressed through an emergency such as the global Covid-19 pandemic.

Progressive curriculum: teaching about the nature of science
The school science curriculum is organised and understood in somewhat different ways in different parts of the world (Taber & Vong, 2020). There have been various arguments about whether or when (i.e., for which age groups) science could be taught as a single subject ('general' , 'coordinated' or 'integrated') rather than as discrete school subjects representing different scientific disciplines (Jenkins, 2007). In the United States, it is quite common for earth science to be seen as a major division of school science alongside the biology, chemistry and physics that have long been seen as the main school science subjects in some other countries (Orion et al., 1999). Astronomy has been taught in some schools. Psychology has sometimes been accepted as a school science subject and, in some countries, geography is seen as a science (although, of course, we should be careful not to assume that labels such as 'geography' are understood to cover the same range of content everywhere).
At one time, in English schools, it was possible to take examination courses in subjects such as rural studies or engineering sciences. In some national contexts, mathematics has been seen as a science subject. In many parts of the world, there has been a focus on 'STEM' (science-technology-engineeringmathematics) or related notions (Chesky & Wolfmeyer, 2015), such as 'STEAM' incorporating agriculture (Sumida, 2018), as a curriculum area, whether seen as a higher-level subsuming category (within which science, or the sciences, will still be discretely taught) or a better focus for the school subject itself.
In part, the discussion behind the merits of making these different choices has been about the scope of natural science to be included in the school curriculum; but, clearly, another issue when considering (i) whether to combine or separate sciences or (ii) whether to form a unitary school subject of science with mathematics and technology, concerns what is common across the sciences. Whereas decisions about how much space science or earth science to include in school science are questions about disciplinary science content (i.e., the products of scientific activity), there has increasingly been a complementary focus on scientific processes. Put simply, this reflects the question of to what extent should school science education be about learning about some of the 'products' of professional science (the theories, the models, the laws, the typologies, the catalogues of 'facts ' , etc.), and to what extent should it be about learning about science qua science (e.g., as a set of practices within a professional community).

The complementary aims of education
There are various potential aims of school education, including facilitating progression to further education and employment; the development of generic areas of skill (such as critical thinking, problem-solving, creativity); the introduction to the key cultural domains valued by the society; supporting personal growth (cognitive, conceptual, ethical, physical, spiritual, etc.) of young people into happy and healthy adult individuals; and the production of citizens prepared to engage in the civil society (for example as voters or as responsible and informed consumers).
Curriculum choices should sensibly be informed by how these competing aims are prioritised. For example, a decision to pack the science curriculum with as much content as possible probably only makes good sense in terms of a focus on progression to higher education; and then, only for those competing for admission to tertiary level science-based courses; and even then, only as long as universities prescribe admissions requirements based on such a breadth of coverage in the curriculum. An in-depth focus on fewer topics might better support intellectual development by allowing greater engagement and more sophisticated treatment of topics; giving emphasis to the needs for informed citizens might also suggest a greater focus on a more select group of topics chosen in relation to societal priorities (e.g., healthy living, the environment, the climate, sustainability).

The NOS turn in science education
It has been widely suggested that the school science curriculum should focus more on what is often known as the nature of science or NOS (Allchin, 2013;Clough & Olson, 2008;Driver et al., 1996). Young people need to understand what science is and 'how it works' (Toplis, 2011), as this will be important for both the minority who become scientists as well as the rest who will engage with science as non-professionals who will vote, spend, recycle (or not), choose (e.g., medical treatments), and so forth in situations impacted by science.
NOS is contested, and scholarly accounts are subtle and nuanced, but there is a general consensus on key features that should be represented in school science (Lederman & Lederman, 2014). Just as many science topics traditionally taught have to be modelled and simplified in the curriculum to be suitable for presentation to school-age learners, curricular models of NOS can be developed (Taber, 2008). There is extensive literature about these issues, but here I offer one illustration.

The key topic of scientific knowledge
One of the biggest challenges for school science teachers is to offer learners a sense of the nature of scientific knowledge, which is largely conceptual and theoretical -and a key principle is that strictly it is always provisional. In principle, all scientific findings are open to being challenged in the future in the light of new evidence or new ways of thinking about the existing evidence (the Copernican revolution and Einstein's ideas about relativity were new ways of thinking that did not depend on any new data). However, we also want learners to appreciate that science is the most reliable means of learning about the natural world and that scientific knowledge is often a good guide to action. For example, Newton's laws of motion are rightly lauded as a major scientific achievement and are still taught in schools today. They were widely considered definitive knowledge for two centuries, although we now know they are, strictly speaking, false (yet under most circumstances work well enough, e.g., in the calculations that allowed people to get to the moon and back safely).
The nature of scientific knowledge is not an easy topic to teach to school children -it is an aspect of the philosophy of science. However, if we want young people to understand, as one critical example, the nature of climate science and public policy debate about climate change, then this becomes essential. Science offers a strong consensus on the effects of anthropogenic inputs into the atmosphere, albeit a small minority of scientists do not accept that consensus. The best scientific models offer predictions, yet these are necessarily imprecise and probabilistic and are regularly revised, suggesting earlier versions were not quite right. It is easy for the layperson to listen to the scientific climate heretics, look at the imprecision and updating of predictions, and conclude that science does not yet 'know' and that we might best defer action until the scientific knowledge is definitive. So, children need to understand that provisional, theoretical knowledge is all we will ever have, and waiting until we know (with absolute certainty) before acting on the science is illogical and dangerous.

Learners should not believe scientific knowledge
It is also useful for teachers to keep in mind that if scientific knowledge is always conjectural and provisional, then it is not their role to ask learners to believe in it. Many people will have learnt scientific ideas at school that have since been demoted from the scientific canon. Science offers us useful ways to understand the world but not an absolute, eternal account. So teachers should ask learners to understand why an idea is useful and why scientists came to suggest it (i.e., in terms of evidence and arguments) but not to believe in the idea (Taber, 2017). As an example, it may be appropriate to teach that general relativity is the best currently available approach to understanding gravity, but it is not in the spirit of science to ask students to believe in the theory of general relativity. Similarly, teaching the 'lock-and-key' model of enzymes and substrates may be sensible as a useful way to think about enzymatic specificity, but it does not make sense to ask students to believe the model. Asking learners to believe in such things would reflect a category error as theories and models are not the kind of entities where belief-disbelief strictly applies, unlike factual claims about what is the case which can be considered to have truth values (e.g., the claim 'Slovenia is a monarchy' would be false).
Science education should include a focus on science as producing models and theories that are often useful in limited ranges of application (e.g., the ideal gas equation) and have to be developed further before they can be applied more precisely or more widely. This would avoid a student, for example, learning a shell model of the atom as some kind of absolute truth, and then finding they are being asked to move beyond this and learn a different account (also just a model, and not an absolute truth): something that can be experienced as having been taught something 'wrong' which now needs to be 'unlearnt' .
Moreover, teachers in many contexts find they are teaching students who, for cultural and religious reasons, are committed to 'truths' that are inconsistent with some scientific ideas. The paradigm case here would be the rejection of macroevolution by natural selection by those who consider that their faith requires them to believe in the discrete special creations of different types of animal and plant groups (Reiss, 2008). Teachers cannot avoid the contradictions between these two perspectives (without abdicating their responsibility to teach the science, cf. Long, 2011), but there is a big difference between asking learners (a) to believe in macroevolution (which logically requires rejecting their faith) and (b) to understand the theory and appreciate the grounds on which it was suggested and why it has become the key organising idea in modern biology. The intellectual clash of ideas is just as great, but without asking for a commitment to a scientific theory as if it was a creed. (Just as in other areas of the curriculum the same students might be asked to understand the viewpoint and actions of a historical figure or of a fictitious protagonist of a novel without being asked to commit to their beliefs, views, or choices.) The increased focus on the teaching of NOS may, inter alia, include more emphasis on enquiry, including historical case studies to show how scientific advances may be difficult and contested, rather than just the retrospective, whiggish, teaching of what has been called a 'rhetoric of conclusions' (Schwab, 1958, p. 375); and engaging with socio-scientific issues (Sadler, 2011) where science can inform social policy, but where decision-making also depends upon consideration of extra-scientific values (e.g., science might quantify the risks associated with building a nuclear waste storage facility or the cost of setting aside an area to protect at-risk species, but cannot tell society how much risk is acceptable, or what cost is worth paying).
It is widely recognised that there can be a considerable lag between the changing of a formal curriculum in terms of documentation and the full acceptance and enactment of the reforms (Peskova et al., 2019). The degree to which aspects of NOS have been incorporated into curriculum and teaching standards and have become part of local custom and practice varies internationally. In many places, this is still progressive and not yet a robust feature of teaching. Indeed, in the English curriculum context, contra international trends, NOS was de-emphasised in the most recent curriculum revision (Brock & Taber, 2019).
It can be considered 'challenging' for many reasons, including (a) the teacher's own scientific education is often lacking in NOS; (b) in many countries high quality texts and teaching resources have not yet been developed to support this area of teaching; (c) teaching approaches may require different pedagogy and teaching skills from those most science teachers have mastered. For example, neutral chairing of a debate about a socio-scientific issue is quite different from teaching an area of established content; engaging with historical sources requires an interpretive approach open to multiple viewpoints, which is not the way science is usually taught. In many national contexts, teaching NOS is 'difficult' from the teacher perspective, and so is a 'fragile' aspect of the practice (cf. Figure 1). When under the stresses resulting from a crisis, it seems inevitable that there will be a reversion to focusing on teaching specific science topics for many teachers, so learning about NOS will suffer. That is, a reasonable hypothesis is that in some educational contexts, curriculum revisions to put more emphasis on learning about the nature of science may lack the resilience to be maintained during a period of systemic stress (and so it is likely that teaching about NOS was less extensive in these contexts during the year 2020 when the global Covid-19 pandemic disrupted education norms).

Progressive pedagogy: taking learners' conceptions into account
The other example I wish to highlight is teaching that takes into account learners' conceptions. The educational psychologist David Ausubel (1968) famously suggested that if he had to reduce the whole of educational psychology to one principle, it would be to find out what the learner already knew -and teach accordingly. This resonates in science education, where much research has highlighted how students commonly form alternative conceptions ('misconceptions') in science topics (Driver et al., 2013). Learners often come to school already having their proto-concepts about natural phenomena, and teaching is often either resisted due to being inconsistent with or inadvertently misinterpreted to fit with prior understandings (Gilbert et al., 1982). Commonly, teachers have to reshape learners' initial thinking, to challenge some alternative conceptions, and to find ways to constructively build upon learner intuitions to channel thinking in the desired directions (Driver & Oldham, 1986).
Again, there is vast literature regarding this (Taber, 2009), and it is not possible to do justice to this area of work here. There are various teaching schemes and particular techniques that have been recommended for teachers.
A key feature of the kind of teaching needed, which might be called constructivist teaching, is interactivity (Taber, 2018a). It starts with (à la Ausubel) diagnostic assessment to identify the students' current thinking. The teacher then seeks to persuade learners towards the scientific view, not simply by presenting that view but through demonstration, argument, discussion, metaphor, analogy, modelling, and other techniques (Hadžibegović & Sliško, 2013;Kress et al., 2001;Lemke, 1990;Mortimer & Scott, 2003).
Most importantly, the teacher constantly uses formative assessment to check how teaching is being understood, checking 'where is student thinking now?' The teaching needs to be dialogic (Mercer, 1995), meaning to have the form of a conversation where the learners' voices are heard. This has often been misunderstood as some kind of relativistic notion that all ideas are equally valued. The teacher does value the students' ideas but not because they are as worthy as scientific accounts, but because learning is always interpretive, incremental, and thus iterative (Taber, 2014), and the students' current thinking is the 'material' available to be worked with to bring about learning and conceptual change.
Again, this kind of approach has been adopted to varying extents in different places. In some parts of the world, the basic principles behind this type of science pedagogy have been reflected in teacher education, curriculum reforms, and official teacher guidance for some years. Effective practitioners present the scientific accounts, but as part of a choreographed practice of eliciting, reflecting, discussing, and challenging students' ideas, and giving learners frequent opportunities to reflect on and work with the ideas the teacher is presenting (Mortimer & Scott, 2003). This kind of teaching is, by its nature, conversational. It is like a symphony, shifting between themes (the received account, the different student notions) and shifting between different solo instruments and ensemble playing (teacher exposition, class discussion, individual reflection, paired and small group discussion).
Teacher talk is not all one-way: it is rich in questions and invitations for suggestions in order to ensure everyone is following, everyone understands, and everyone's ideas are getting a hearing. All ideas (whatever the source) are open to communal critique in terms of logic, evidence, argument structure, and coherence with other ideas we accept. This also models the core scientific value of questioning and testing all contributions on their merits. If this teaching style becomes too difficult, this means a less effective way of teaching science concepts and also the loss of an implicit way of reinforcing a key feature of NOS.
Again, in a time of great stress on schooling and teachers, it is likely that those practitioners who are less experienced at these techniques, where such practice is still 'fragile' , will readily slip back to 'teaching by telling' . Moreover, it seems likely here that even those teachers who have mastered such approaches and have made them part of their normal custom and practice (such that they can be considered 'robust' rather than 'fragile') may be challenged to teach in this way when faced with a class as a set of tiny muted headshots on a computer screen. That is, a reasonable hypothesis is that in some educational contexts, pedagogic reforms to better support student construction of knowledge through dialogic teaching may lack the resilience to be maintained during a period of systemic stress (so in these contexts it is likely that science teaching tended to revert to direct communication of the 'received' account during the year 2020 when the global COVID-19 pandemic disrupted education norms).
Perhaps, with the right technology, and time to test out teaching methods, it will prove just as effective to teach science, taking into account learners' ideas, via the Internet as it is in the classroom (Taber & Li, 2021). The use of chat rooms and the like can substitute for breaking the class into small groups for face to face discussion (and without groups distracting, or 'borrowing' from, each other). Wikis or shared glossary tools may be used to collect different learners' ideas and suggestions simultaneously, and possibly more effectively, rather than sequentially asking each learner or group in a classroom. However, even if that is true in principle, it will not be a straightforward transition but rather something that will require development and practise, just as any 'reform' does. So, it may be that teaching virtually is not in itself the challenge, but rather the sudden shift between classroom and virtual teaching without suitable warning and preparation. It is also possible, however, that distance learning (with the technology available today, at least) simply does not lend itself to effective science teaching as well as the classroom.

Conclusion
This article makes an argument that the stresses placed on the school system during the Covid-19 pandemic will inevitably impact the quality of the teaching and so student learning, and that this will disproportionally affect those aspects of teaching which might be seen as desirable but not essential to 'delivering' the curriculum, and which are felt more 'difficult' and so need to be put aside when seeking to 'make do' and 'get through' in a crisis. Wellestablished aspects of custom and practice are likely to be robust features of teacher practice, whereas elements associated with 'reform' and thus still seen as progressive are more 'fragile' and subject to being given a lower priority. An obvious challenge to science teaching in lock-down conditions is practical laboratory work which, despite being a robust aspect of science teacher practice in most countries, presents major logistical challenges to moving online.
In this article, I have, however, focused on two other areas where I predict science teaching quality will have suffered, two areas that have over many years been much discussed in the literature and which have to varying degrees been adopted as aspects of educational reform in many national contexts. One prediction is that teaching about NOS will have suffered more than teaching science content in those contexts where teachers still find this a more challenging and/or peripheral aspect of their work. The other prediction is that the kind of dialogic teaching at the core of constructivist approaches which take into account learners' ideas, which is seen as critical to effective teaching of science concepts, and which relies upon teachers' interpersonal skills in making science lessons more like conversations than lectures, will prove more difficult online. For some science teachers, this will still be seen as a 'desirable' rather than 'necessary' aspect of their work, but even where this approach is well-established and so not as inherently fragile, the online mode is likely to encourage a shift back to teaching that is based more on a telling of the canonical account.
Hopefully, in time, there will be studies that explore the extent and nature of changes to teaching during the Covid-19 pandemic, and such research will help education systems become more robust in preparation for future crises that might require similar sudden changes in the organisation of teaching. If the findings of empirical work reflect the predictions made here, then part of that preparatory work should involve considering how one protects progressive elements of educational policy and practice in such circumstances. After all, reforms are made to improve teaching and learning, and so it is important to mitigate the fragility of those elements and seek 'reform resilience' in the face of stresses to the educational system.