Professional Judgement and Standards in Educational Assessment

Note: This is a slightly adapted version of a lecture given to Adult Literacy Organisers in Galway, Republic of Ireland in April 2013, under the title: Initial Assessment - What Works? The lecture was organised by The Republic of Ireland's National Adult Literacy Agency (NALA). A video of part of the lecture is available here.

I’m here to talk about Professional Judgement and Standards in Assessment.  Most of what I want to say doesn’t just apply to initial diagnostic assessment in Adult Literacy work, but to teaching in general, and to many other areas of human activity.  It builds on a paper written for my doctorate (Derrick 2013).

I want to start by giving you all some questions to think about.  How much experience do you have as an adult literacy practitioner – hands up: more than ten years?  Five to ten years? Under five years?  Anyone in their first year?  Can I ask the people with more than ten years’ experience – would you say you are good at Initial Diagnostic Assessment?   What is it that makes someone good at it?  For example, is experience essential?  How important is formal training?  How can you learn to get better at it?  Is there a formula, or a recipe for getting better at it?  Is Initial Diagnostic Assessment a science?  Is there, or could there be, an instruction manual for it?

The answers to these questions are vital for sorting out the role of standards, skills frameworks, levels, etc in the work of adult literacy teaching.

The three main propositions of my talk are these:

• Expert teaching (and within this, expert assessment) is essentially about the exercise of professional judgement in making decisions
• Professional judgement can be supported by all sorts of tools, resources, measurements, rubrics, standards, etc, but cannot be replaced by these
• Learning to become better at teaching and assessment is a process of training professional judgement.

30 years ago, Donald Schön, writing about professional practice in general, said this:

In the varied topography of professional practice, there is a high, hard ground where practitioners can make effective use of research-based theory and technique, and there is the swampy lowland where situations are  confusing ‘messes’ incapable of technical solution.  The difficulty is that problems of high ground, however great their technical interest, are often relatively unimportant to clients or the larger society, while in the swamp are the problems of the greatest human concern. (Schön 1983 (1991), p42)

Schön suggests that professionals need to be able to deal with, not just technical, generic problems for which there is a recognised procedure, a standardised routine, but also other, more complex, essentially unique problems for which there is, and can be, no standard ‘correct’ response, and which are therefore amenable only to judgement calls. 

A practical example of this type of problem might be the question of whether to take a child into local authority care, which has to be considered all too often by social workers.  There can be no absolute instruction manual for making a decision like this, because every situation is uniquely different.   

Standard procedures, checklists, etc are helpful, but in the end a decision like this is not reducible to standard procedures – we can only try to ensure that they are made by people highly motivated to make the best possible decision, and who have as much knowledge as possible of the particular situation – and note that ideally this knowledge is not just technical knowledge of the law, of child and of family psychology, etc, but also personal knowledge of the unique pattern of the individual circumstances of the people involved.

Schön’s work is a sustained argument against the dominance of what he calls the ‘technical-rational’ view of professional practice.

Procedures and rubrics are models or hypotheses which always simplify and ‘clean-up’ the ‘messiness’ of real-life situations.  As Goethe warned:

‘Hypotheses are scaffoldings erected in front of a building and then dismantled when the work is finished.  They are indispensable for a workman, but you mustn’t mistake the scaffolding for the building.’ (Goethe Posthumous (1998), p154)

The scaffolding is useful, even perhaps essential, but it cannot replace the building which it supports.  This it seems to me, is an accurate metaphor for the role of rubrics, standards, technical terminology, pro-formas, skills frameworks, assessment criteria, grade descriptors, diagnostic assessment tests and procedures, and the like.  They have their uses, but ultimately teachers need to make the decisions, because each situation, each learner, is essentially unique.  This is why we should resist talking about education as if it was like mass-production – it is not about using standardised raw materials to produce identical objects.

Addressing educational problems as if they can be reduced to a technical model of one kind or another ignores the local knowledge, practical wisdom and perhaps the long experience of the professional in making similar decisions.  It also ignores the moral dimension: that the professional is motivated to find the right way forward for the clients and others concerned – by which I mean the best and most just response to the situation, not just a technical solution which ‘will do’.   A hypothetical model cannot be relied on to do this because it is a simplification of the situation, and by definition ignores factors in it which in a particular case may be critical.

The holistic argument I am making here highlights the importance in decision-making of ‘local distinctiveness’: a nuanced awareness of what is distinctive in the features of a particular locality or situation.  Schön might put it as appreciation of the distinctiveness of each particular ‘swamp’, so as to make informed decisions about appropriate actions.  This view is a critique of the kind of overly-simplistic ‘technical-rational’ modelling of situations that  makes decision-making (for example about whether to build a new supermarket in a country town) appear to be easier, perhaps to the long-term detriment of the locality and its community.  Sue Clifford and Angela King (2009) argue against the homogenising tendency of technical-rational modelling in all aspects of life, emphasising the importance of detail, particularity, patina, authenticity, fineness of grain, divergence, messiness, and quality in the everyday.  This may be caricatured as a conservative critique of regulation through standardisation, of technocratic policies and procedures: but it is an argument primarily against shoddiness, against work in any field that will merely do, rather than reflecting the best efforts of the practitioner.  Research (eg Black and Wiliam 1998, Hattie 2009) tells us clearly, for example, that teachers’ sense of the particularity of situations is central to their understanding of the specific needs of individual students at a given moment in the learning process: this sensitivity is also undoubtedly a foundation of trust and mutual respect between learner and teacher.  It is clearly possible to teach without this detailed knowledge of students and their situation, and perhaps even to teach ‘adequately’, but professional practitioners are deeply and morally concerned about ‘local distinctiveness’ as part of doing the job as well as possible.  Attempting to encapsulate this sensibility in a rubric will always tend to diminish it: minimum standards tend to produce work that simply satisfies the minimum acceptable level of quality. 

In England, much has been made in recent years about the use by teachers of data in their decision-making: it has been suggested, perhaps with some justice, that they have relied too much on hunches and on their intuition, and not enough on ‘objective’ data about their students’ progress.   There is also evidence that teachers and other professionals have sometimes been discriminatory in their practice (for example, having lower expectations of students from particular backgrounds), and this too has been used to support arguments that their work should be regulated within procedural and externally-imposed quality standards.

In my view this is to throw the baby out with the bathwater: it is to argue that the work of teachers is simple rather than complex; it is to argue for the production of an instruction manual for teaching, one consisting of a highly intricate flow-chart, that would provide teachers, whether qualified or not, with a procedure for correctly addressing every teaching problem, or, in terms of our topic today, for correctly the assessing the skills of every new student.

I think that most people would agree that this is not what teaching is about: they recognise that the work of teachers is essentially complex and cannot be reduced in this way.  On the contrary, teachers should be supported to develop their intuition and their capacity to make good judgements, just as surgeons, doctors and social workers do.  So what is the role of all these frameworks, rubrics, standards, pro-formas, tests, etc? It must be to provide information that teachers can use in their decision-making, and in discussing with students their learning goals, their attainments, feedback on their work, and, most importantly, how they can get better at supporting their own learning – in a word, these tools can be used to support formative approaches to learning, provided the information they provide supports, rather than determines, the authentic judgements and feedback of the teacher.  In this way, tools can also support teachers’ own professional learning.

From this perspective, it is not the tools themselves that are good or bad, it is how the teacher uses them that matters.  It is important, though, that teachers are free to determine for themselves how best to use them.  If they are not free in this way, then a particular technical-rational model is being imposed on them, and this may lead to students being sold short, and also to the deskilling and demotivation of teachers.  The problem for policymakers is that education is full of swampy, messy situations, which by definition are hard to predict, control and regulate.  To make the best of them we need teachers who are confident in weighing up situations, highly effective communicators in terms of knowing the kinds of questions to ask and how to listen, expert at qualitatively judging students’ strengths and weaknesses, and motivated to make the best decisions possible about the right courses of action for the student and for the next stage of teaching and learning.  They are very likely to find supporting technical information useful in making these judgments and decisions, and they should certainly pay attention to such data.

It is difficult for policymakers to regulate teachers without imposing limitations on their professional autonomy which may limit their scope for action, as well as reducing their motivation.  But this difficulty does not justify the misconception of learning as merely a technical issue of standard procedures, and teaching as a simple, technical job involving the obeying of standard rules.

If we had time I would invite you all to consider how best we should organise the training and development of adult literacy teachers in a non-technocratic policy environment.  Instead I’ll leave you with an inspiring story of exemplary professionalism that illustrates many of the issues we have been addressing here: on January 15th 2009, a commercial aircraft with 150 passengers and crew on board took off from La Guardia airport in New York.  The pilot in command was 57-year-old Capt. Chesley B. "Sully" Sullenberger, a former fighter pilot who had been an airline pilot since leaving the United States Air Force in 1980, and who is also a safety expert.  During the plane’s initial climb, it struck a flock of Canada geese. The bird strike, which occurred about three minutes into the flight, resulted in an immediate and complete loss of thrust from both engines. When the crew of the aircraft determined that they would be unable to reliably reach any airfield, they turned southbound and glided over the Hudson, finally ditching the airliner in the river about three minutes after losing power.  All 155 occupants safely evacuated the airliner, which was still virtually intact though partially submerged and slowly sinking, and were quickly rescued by nearby ferries and other watercraft.  The recording of the incident makes it clear that, immediately following the bird strike, it was obvious the plane had to land as soon as possible, and there was intense radio communication from the air traffic controllers, concerning availability of nearby runways, and the distance to each one.  After about a minute of listening to this information, the pilot asked just one question about the possibility of landing at New Jersey but then suddenly seems to make up his mind: ‘We are gonna be in the Hudson.’  He then said nothing more, apart from ‘Brace for impact’, until the plane was down.  In interviews afterwards he said that there were two factors that helped him pull off the manoeuvre safely: that the plane’s design was state of the art, and that he himself was 55 rather than 25.  ‘In a sense my whole career had been training for this event’ (Couric 2009).

This story illustrates how all sorts of tools, procedures and data made important contributions to the successful outcome of this manoeuvre, but these would have been irrelevant without the skills, experience and professional behaviour and attitudes of Captain Sullenberger and his crew.  None of these guaranteed success, because off the technocratic map (ie, when the unexpected happens) success is never guaranteed.  The experienced and motivated professional aims to optimise the chances of success.  Even though teachers rarely deal in life or death decisions like the ones Sullenberger found himself taking, in reality the various parts of their job, including initial assessment, are more often off the technocratic map than on it; and this means that the question ‘What Works?’ is misconceived. 

References

Black, P. and Wiliam, D. (1998).  ‘Assessment and classroom learning: Principles, policy and practice’. Assessment in Education 5 (1), pp 1-78

Clifford, S. and King, A. (1993: 2009 edition). Losing Your Place, in Local Distinctiveness – Place, Particularity, Identity.  Accessed 05-04-13 at http://www.aughty.org/pdf/losing_place.pdf 

Couric K (2009).  Interview with Chesley Sullenberger accessed 05-04-13 at http://www.youtube.com/watch?v=9qKXgG-imh4

Derrick, J. (2013). ‘Practitioner learning’, not ‘professional learning’: towards a non-technocratic model of autonomous development for post-compulsory sector teachers.  Research in Post-Compulsory Education. 18 (3) pp269-279

Goethe, J.W. (Posthumous, 1998 edition). Maxims and Reflections, ed P. Hutchinson . London: Penguin

Hattie, J. (2009). Visible Learning. Abingdon: Routledge

Schön, D. (1983, 1991 edition). The Reflective Practitioner: how professionals think in action.  Aldershot: Ashgate Publishing Ltd

Cultivating and developing intuitive decision-making in teaching

I have to admit that throughout most of my career my ideas about, and practice in, effective teaching have been relatively incoherent and intuitively arrived at, rather than the products of careful and analytical thought. According to Yvonne Hillier (1998) this is common even amongst the most experienced and well-trained of teachers.


At the same time, I don't subscribe to the view that deliberate reflective practice is at the opposite end of the spectrum from intuition: I firmly believe that valuable insights can be gained from both rational thinking and from inspiration, and that in principle it is valuable for professionals in any practice to be receptive to insights that come unexpectedly, unplanned, and from unlikely sources, as much as those that are the product of rational processes of deliberation and analysis. I posted a couple of months ago about the value of deliberately going off piste and 'seeing what happens', and this view connects closely with the idea that effective teaching is not necessarily a neat and tidy business (Derrick 2010).


The tension between these different modes of apprehension has often been polarised into antagonistic and supposedly irreconcilable approaches to the practice of teaching (and indeed practice in other domains): I believe rather that these modes of thought should be seen as complementary, and that teachers need to be able to negotiate a balanced track between the two to be really effective.

Highly relevant to this suggestion is Daniel Kahneman's brilliant survey (2011) of psychological research studies into the way we make decisions: It is as if, he argues, that there are two ways in which we make decisions, which he calls Systems 1 and 2.



System 1 is very quick, in fact more or less involuntary. It utilises what might be called ‘intuition’:


We have all heard….stories of expert intuition: the chess master who walks past a street game and announces “White mates in three” without stopping, or the physician who makes a complex diagnosis after a single glance at a patient. Expert intuition strikes us as magical, but it is not. Indeed, each of us performs feats of intuitive expertise many times each day.... Our everyday intuitive abilities are no less marvellous that the striking insights of an experienced firefighter or physician – only more common. The psychology of accurate intuition involves no magic….Valid intuitions develop when experts have learned to recognise familiar elements in a new situation and to act in a manner that is appropriate to it. (Kahneman p11)


System 2 thinking, on the other hand is slow, requires motivation and effort, and uses what we might call ‘rationality’. We use these different systems in different circumstances, confronted with different types of problem. We prefer to use System 1, which evolved as a survival mechanism, and with which we make very quick decisions based on the continually developing capacity for correct intuitive judgements which we have developed throughout the whole of our past lives. Of course these decisions may sometimes be wrong, but in familiar situations System 1 decision-making has evolved to be accurate most of the time, and as a result it takes a great deal of conscious effort to go against what it tells us.


System 2 is used for problems that clearly do not require an immediate solution, and/or which demand that we follow a procedural algorithm in order to solve them, such as a complicated long-multiplication sum. Kahneman argues that we only use System 2 reluctantly, when we have to, because it takes effort and energy, whereas System 1 thinking is effortless and easy. System 1 decision-making is more likely to be accurate the more familiar we are with the situation we are in; it can work astonishingly well in such familiar situations, even if the problems involved are very complex, indeed too complex to be easily solved by the use of System 2 heuristics and algorithms.




This research-based psychological typology of our decision-making capacity is strikingly reminiscent of Donald Schön’s well-known theoretical distinction between reflection-in-action and reflection-on-action (1983). The key issue for practitioner learning, therefore, given the need for expertise in tackling both well-defined ‘high ground’ problems and complex, ill-defined, ‘swampy’ problems, is how best to develop and improve ‘reflection-in-action’ or System 1. For a new idea to contribute to changed System 1 decision-making, according to Kahneman, and in different terminology, to Schön, the practitioner has to 'practice' using it, reflect on this practice consciously and probably collaboratively (System 2 activity), and repeat this many times over a long period. 'The accurate intuitions of experts are better explained by the effects of prolonged practice than by heuristics' (Kahneman 2011, p 11): there are no short cuts, no quick intellectual fixes or magic bullets, in shaping our intuitive capacities.



This System 2 work of developing System 1, essential to practitioner learning, is difficult, both to carry out and to reflect on: it takes a great deal of conscious effort because it involves a decision to engage in practice in a different way from that informed by our System 1 thinking. This demanding process is one way to imagine that practitioners might strive to enhance their intuitive capacity for ‘divergent thinking’, cited by Schön as an essential capacity for tackling ‘swampy’ problems. Such careful, effortful, disciplined and probably repetitive practice over time does lead to changes in patterning and routines in relation to our work, which gradually influence changes in our intuitive responses to it, and so to our intuitive judgements and actions. What is needed as part of practitioner learning, therefore, is to support a deliberate process of learning that mimics the way our intuitive decision-making capacities have unconsciously and continuously developed and evolved, since we used them to save ourselves from being eaten by lions on the plains of Africa millennia ago.


Derrick J (2010): ‘The messiness of real teaching and learning’, in J. Derrick, U. Howard, J. Field, P. Lavender, S. Meyer, E.N. von Rein, and T. Schuller (Eds), Remaking Adult Learning: Essays on adult education in honour of Alan Tuckett. London: Institute of Education.
Hillier Y (1998): Informal practitioner theory: eliciting the implicit.  Studies in the Education of Adults, 30 (1) 35-52
Kahneman D (2011): Thinking, fast and slow. London: Allen Lane
Schon D (1983): The reflective Practitioner: how professionals think in action.  Aldershot: Ashgate Publishing Ltd

Standardisation in the work of Teaching and Learning

Forcing everyone to teach by the rules does not improve the “bad teachers”– it only hobbles the good ones (Papert 1990)

I came across this quote from Seymour Papert in Joe Bower's post at http://coopcatalyst.wordpress.com/2011/03/13/the-illusion-of-standardization/. He sees standardisation in education as one of the ways in which mass production techniques are being applied to the work of teachers. The justification for this is 'fairness': the argument is that everyone should have an equal experience of their education, otherwise something isn't fair. The fact that the quality of an experience - how it is felt by two different people - is completely immeasurable, and actually is a completely meaningless idea, is forgotten because of our fear of being unfair. Sadly, the price for alleviating this fear is 'dumbing down' to the minimum level set by the standards, exactly as Papert pointed out more than 20 years ago. Supporters of standards argue that because there are so many bad teachers, we have to have standards to ensure that teaching quality is consistent. This is another concept, along with standardisation, that needs looking at carefully by every teacher.

Joe also includes a link to an article by Maja Wilson, who tells the story of her son choosing a book to read and taking it to school, but being told off because he hadn't chosen the book from the 'levelled book tubs'.

'Standardisation' in education is a paradigm indicator of what Donald Schon, nearly 30 years ago, described as 'technical-rationality':

In the varied topography of professional practice, there is a high, hard ground where practitioners can make effective use of research-based theory and technique, and there is the swampy lowland where situations are confusing ‘messes’ incapable of technical solution. The difficulty is that problems of high ground, however great their technical interest, are often relatively unimportant to clients or the larger society, while in the swamp are the problems of the greatest human concern (Schön 1983 (1991), p42).
Schön uses the term ‘technical rationality’ for what he describes as


the dominant epistemology of practice...which has most powerfully shaped both our thinking about the professions and the institutional relations of research, education and practice – professional activity consists in instrumental problem solving made rigorous by the application of scientific theory and technique (Schön 1983, p21).

He presents a compelling, almost overwhelming argument that the dominance of this version of epistemology in the theories of learning, research and professional practice has obscured the real importance of alternative, more expansive and provisional knowledges, resulting in the diminution of professional expertise and creativity in the design of a wide range of projects, and in addressing major social and educational problems. Unfortunately, his work appeared in the period just before two momentous shifts in Western politics and culture: the advent of computers on every workdesk, and the collapse of the Soviet Bloc. These shifts gave a renewed and powerful impetus first to neo-liberal economic policies and secondly to a new wave of technical-rational social policy fuelled by the new capacity of organisations both public and private, to collect and crunch data. Arguments such as Schön’s for a more nuanced and complex appreciation of the work of professionals of all kinds were swept aside in the rush towards the ‘Audit Society’ – a paradigm of technical rationality, ‘audit’ being seen in strictly well-defined, contractual and numerical terms (Power 1999). This mode of thought is still dominant, probably to a greater degree than when Schön’s work was published nearly 30 years ago, and professional practice is even more clearly subject, though not unresisting, to its sway, not just in the structure, organisation and curricula of universities, as Schön demonstrated in 1983, but nowadays also in social, educational and economic policy, at least in the UK and the USA, where it has become an explicit element of the philosophies of every major political party.

As a result of this continuing dominance, the situation of professionals in every field has become much more problematic than when Schön was writing. They are far less trusted, they are less autonomous in practice, early ideals have been watered down, they are subject to various versions of ‘performativity’, the ethical issues they face are far more complex.  But the most important consequence of this situation is that the work of dedicated, thoughtful and hard-working teachers is undermined, with obvious consequences for learners of all kinds.

Papert. S. (1990, July). Perestroika and Epistemological Politics. Speech presented at the World Conference on Computers in Education. Sydney, Australia

Power, M. (1999). The Audit Society: Rituals of Verification. Oxford: Oxford University Press

Schön, D. (1983, 1991 edition). The reflective Practitioner: how professionals think in action. Aldershot: Ashgate Publishing Ltd

Professional learning by deliberately cultivating the unexpected

“He that confines himself to one book at a time, may be amused, but is no student.  In order to study, I must sit in some measure in the middle of a library.”

William Godwin, The Scots Magazine and Edinburgh Literary Miscellany, Vol 81, 1818

I take this to suggest that 'students' need to do more than focus on and work directly at the subject of their studies; they must enrich this direct approach with other ideas, stimuli, experiences, so as to allow and encourage surprises, unexpected insights; in short, serendipity.  This might be boringly described as 'reading around your subject', but I think Godwin means something much more than that: something more like Rimbaud's 'systematic derangement of the senses', so as to try to be jolted out of one's preconceptions. 

I am pleased with this idea, as it aligns well with something I wrote recently about the messiness of real teaching and learning (Derrick 2010):

In the dominant model of public sector professionalism today, accidents are seen as prima facie evidence of failure: if unexpected events do occur, it must be due to inadequate preparation or human error. This view of practice assumes that in principle the job and skills of a teacher, social worker, parent, etc, can be defined in detail, without ambiguity, and therefore codified precisely. Regulation and quality assurance of these roles, it follows, is a straightforward process of checking that there is no variation from the code. The training of such professionals is also seen as straightforward, based on a well-defined, unproblematic body of knowledge and an accompanying skill-set that hardly changes over time. This positivist view of practice derives from a positivist view of learning, in some versions almost identical to behaviourism, in which the individual learner is seen as a passive recipient of pre-defined knowledge and skills, which are acquired usually through processes of memorisation and repetitive practice....

There is little room in the positivist view of practice for the idea of teaching as an art, or as a craft, something honed and fashioned over time in the context of a ‘community of practice’, and which takes for granted that changing circumstances and new learners will produce new problems for the teacher, who therefore needs the capacity to respond to new and unexpected situations.  Recent research shows clearly that teachers need to be ready at times to abandon their detailed plans and ‘go with the teachable moment’, not just as a response to difficult situations, but as an optimal strategy when things are anyway going well. Nor can the positivist view encompass the idea of the learning process as active, combining research, creation and re-creation by groups as well as by individuals, with a range of possible outcomes and wider benefits at the level of the individual learner, the group, the community and society in general . It does emphasise the role of the learner in choosing programmes of learning to enrol on, just as a shopper chooses products from the shelves of a supermarket, but this can only be equated with active learning from a reductionist, behaviourist viewpoint....[This] echoes Donald Schön’s famous distinction between different types of professional problem:
In the varied topography of professional practice, there is a high, hard ground where practitioners can make effective use of research-based theory and technique, and there is the swampy lowland where situations are confusing ‘messes’ incapable of technical solution. The difficulty is that problems of high ground, however great their technical interest, are often relatively unimportant to clients or the larger society, while in the swamp are the problems of the greatest human concern.  (Schön 1983)
This view suggests that a crudely positivist philosophy and policy of learning, teaching, and teacher development, whatever its intentions, will be damaging both to society and to individuals. In particular, it will lead to deskilling: it nurtures in professionals a passive, bureaucratic and parochial version of their work, which over time leads to diminishing interest in, curiosity about, and capacity to deal with unexpected situations, or even to imagine crisis scenarios so as to prepare for them. Specifically it leads to diminution in preparedness and of the capacity to make judgements about the best course of action in the difficult, complex situations to be found in classrooms everywhere.
Richard Sennett (1970) argues that everyone needs to experience living and coping with the uncertain and the unfamiliar, of managing and dealing with life in complex and unpredictable situations without being in control of them, in order to reach full psychological maturity. If state-supported education seeks to eliminate the uncertain, the debatable, the unexpected, he suggests that it is condemning people to a kind of passive and frustrating adolescence, whatever their biological age.
Avoiding these problems, and harvesting the full potential of learners and teachers to enrich their practice for the economic, social and cultural benefit of society in general, will mean taking a quite different approach to learning in general, and following its implications through in terms of curriculum, the organisation, funding and accountability systems for state-supported learning, and of course to the formation and ongoing development of teachers.
References:

Derrick J (2010) The messiness of real teaching and learning, in Derrick J, Howard U, Field J, Lavender P, Meyer S, von Rein EN, and Schuller T (eds, 2010) Remaking Adult Learning: Essays on adult education in honour of Alan Tuckett.  London: Institute of Education

Schön D (1983) The reflective Practitioner: how professionals think in action. London: Maurice Temple Smith
Sennett R (1970 - 2008 edition) The uses of disorder – personal identity and city life. New Haven and London: Yale University Press