Global inscription devices and the politics of education.
In recent years, a range of indicators, ratings, rankings, scales, league tables, tests, statistics and performance measurements have made themselves increasingly relevant in educational policy. PISA is just one example of a quantification regime that is large-scale, institutionally embedded, and which exerts powerful influence upon educational policies including the epistemic level concerning the very definition of education and its purposes (Seller and Lingard, 2013). In this article, we shall take a closer look at a larger category into which this phenomenon falls, i.e. 'Inscription devices', a term for methodological, statistical, organizational and practical tools that render visible and transform otherwise complex, ambiguous realities into figures, scales, indicators, numbers or categories that in turn make possible comparison, rating or ranking.
Numbers have played a strong role in the history of civilizations (Cohen, 2005). While the implication of statistics for the exercise of power and for the production of narratives and meaning at least dates back to year 0 (the basis of all numbers!) when Virgin Mary and her Joseph were on the way to participate in a census (an inscription device!), the systematic and institutionalized production of digital statistics on the global scale is new, and so is the intensification of use of this phenomenon in educational policy. In many countries, reference to PISA scores and other international comparative data seem to be the central driver of reforms--even if it is also a central issue in controversies about educational policy. Although the consequences of global inscription devices may be many--and some perhaps complex, invisible, or contextual-local--the first and most important hypothesis that springs to mind is that global statistical comparison assumes that all countries have the same educational goals. Global inscription devices thus reduce local, practical, normative and political differences and create a commensurate space in which it is pretended that the meaning of education all over the world is one and the same (Meyer, 2008). Once these statistics become seen as valid representations of a reality that should be taken seriously and should be acted upon, then the assumption about isomorphic goals gradually slides into a constitutive idea that influences educational policy in powerful ways.
How is that possible? How is it possible for global inscription devices to play a seemingly central role in the definition of educational goals, especially when they are not bestowed with conventional sources of democratic legitimacy? Parliaments and ministers of education describe PISA results as externally given, as exogenous reality, while teachers and educational researchers are uncomfortable with such statistics and ask for more "debate", although there is little evidence that global inscription devices would really operate differently if more debate took place. We are faced with an interesting paradox: While it is difficult to see the governance of global inscription devices as democratic in any classical sense of the word, they are intensely political in their operations and their effects.
Researchers in recent years have pointed out that this restructuration of educational policy would not have been possible without a fundamental transformation of the social landscape in which educational policy is defined. Classical notions of time, place and linear effect are being replaced by new notions. Instead, heterarchies have multiple centres and consist of heterogeneous multiple elements that are co-functioning through multiple relations (Ball, 2009). Thompson and Cook (2014, p. 1) mobilize the notion of topology to highlight the self-organizing, emergent spaces in which being and knowing are brought into new relations. Appadurai (1996) suggests that globalization should be understood in terms of flows in ethnoscapes, technoscapes, financescapes, mediascapes, and ideoscapes. In that spirit, it would be tempting to suggest that datascapes should be added to the list. Yet, we are lacking a good theory of global inscription devices that would help us understand more specifically how global inscription devices operate. To build this theory is an enormous task. Although broad explanations as "power" and "ideology" should be taken into account in such theory, the two cannot stand alone faced with the complex dispersion of imaginaries, institutions and actors at both global and local levels that weave ideascapes, financescapes, technoscapes, ethnoscapes, mediascapes, and datascapes together and make global inscription devices operationally possible. It is the available forms of social organization that might determine whether it is possible to count at all (Porter, 1995, p. 35), so understanding how a global inscription device can be constructed in the first place may be an important step in understanding how it can work and how it can be powerful.
While the focus in this article is on understanding the impact of global inscription devices upon educational policy, it may be fruitful to go back to fundamental how-questions about the operations of global inscription devices in educational policy as well as to look sideways upon other global inscription devices in other policy areas as well and thereby discover variations that help us build theory.
The strategy of this article is therefore to formulate a handful of questions that a good theory about global inscription devices should be able to answer. I shall argue that in order to formulate these questions, Latour's Actor-Network-Theory is useful as well as consistent in spirit with the heterarchies, topologies and flows conceptualized above.
Before I turn to each of these questions, let me account for the overarching idea that unites them. My overall orientation is constructivist, and on this basis I shall explain the idea of a "contestability differential" which I find particularly helpful for understanding inscription devices.
Finally, I will present a very brief example of how a global inscription device enrolled actors, including myself, in a short media controversy over educational policy in Denmark. Hopefully, that example will demonstrate the usefulness of my five questions, as well as their limitations.
2. Social Constructivism as an Approach to Inscription Devices
Social constructivists are generally interested in how particular social realities become established and taken-for-granted (Berger and Luckmann, 1966). The better social construction works, the more a given social reality is takenfor-granted. Institutions are regulated through norms and values, although tougher regulatory sanctions help keep institutions in place only in more extreme situations. In a socially constructivist perspective, an institution is effective if its framework is uncontestable.
When looking at a particular social phenomenon, we can analytically emphasize that it is constructed, meaning that it is a result of human interaction, language, institutions etc. We can also emphasize how a phenomenon is involved in constructing something, such as a particular institutional order, or, for that matter, a change in values, practices etc. However, we should not be fooled by this analytical distinction. In daily life, the constructed and the constructive aspect of social interaction cannot really be separated. As Latour (2005, p. 176) reminds us, none of the large-scale "structures" that we use as social explanations would explain anything if they were not connected to actors and actions. As Latour's Actor-Network-Theory (ANT) suggests, actors themselves might be of many kinds, material as well as social. Thus, a device that produces inscriptions may be as socially productive as a human being or any other kind of actor. In social construction, you cannot begin by assuming that some building blocks already exist, and then let construction begin. Connections and flows should occupy us, not building blocks and causalities. Recent studies in the sociology of knowledge have shown that the production of a particular piece of knowledge requires a large set of connections between multiple elements in a whole "hinterland" of constructions undergirding an innovation (Bijker and Law, 1992). At the same time, neo-institutional organizational theory, another school of thought in the sociology of organized life, suggests that due to the complexity of fragmented and dynamic institutions that operate in global-local spaces, there will be disconnections between organizations and their environments, between strategies and practices, and between talk and action (Meyer, 2010). In other words, although both have constructivist thinking of flows as a central tenet, ANT puts a relative emphasis on connections, while neo-institutional theory maintains an interest in loose couplings and disconnections. Neither of the theories suggests, of course, that everything is either connected or disconnected, but the conceptual figure-ground-relationship is surely different between these two schools of thought. The theoretical tension between the two may be intellectually productive.
I shall leave a number of nuances and internal discussions among constructivists--of which there are many kinds--aside and go forward to a discussion of the specific characteristics of inscription devices as elements in social construction. While all social constructions are in a "flow", inscription devices and other evaluative tools and procedures such as benchmarking, accreditation, auditing, and evaluation are constructions of a special kind. They are "assisted sense-making" (Mark, Henry and Julnes 2000, p. 5), i.e. they are deliberately constructed with the purpose of facilitating a comparison or analysis that makes it possible and legitimate to act upon the reality they describe. They are devices designed to accelerate social construction. They are built to induce change. In that sense, they are indicative of a modern imaginary that assumes that social reality can be mastered through technological tools (Berger, Berger and Kellner, 1973). As Koselleck (2006) states, concepts and ideas that are particularly modern reveal "open-ended" social expectations. The social order is in principle manipulable. It can and should be changed, modernists believe, although they of course disagree about the specific direction of change.
Statistics have a function that resonates with Koselleck's idea about modern concepts. Numbers are variable by definition; then why can't the social order be changed so that problems are ameliorated and numbers that describe problems change accordingly? This is what we call "social improvement." Theoretically, there is a solidarity between statistical inscription devices, social contingency, and planned social change. Inscription devices are indicative of a society that seeks to live up to an ideal of self-appropriation (Stehr, 2001), a society that wants to deliberately organize itself.
If this is correct, a particular task is placed on the shoulders of any evaluative tool as an element in a socially constructive process. Taken-for-granted-ness should now be attached not to a particular social order, but to a particular kind of intervention or change. An evaluative tool, to be successful, should be aligned with social constructions that are relatively uncontestable, while other constructions, the objects of evaluation, should be made contestable so that they can be changed.
How can an evaluative tool present whatever is under evaluation as something contestable? First of all, the very idea that something should be evaluated is in itself an imposition of a set of meanings upon the evaluation object. It suggests contingency where there was perhaps no contingency before, because objects of evaluation are potentially under change, since evaluation aspires to lead to improvements.
Next, an evaluative tool can enrol a discursive presentation of, say, teaching practices as old-fashioned, incompetent, not competitive, unsystematic, not evidence-based, etc. The more problematic the evaluand is, the greater is the need for evaluation and improvement. On the other hand, the evaluator (understood as a broad Latourian actor composed of multiple modalities such as an institution plus people plus an inscription device) should be understood as less contestable. Evaluative inscription devices should be necessary, trustworthy, scientific, reliable, useful, relevant, authoritative etc. In other words, there should be a contestability differential between the evaluator and the evaluand if an inscription device is to achieve maximum success as part of a construction of a changed social reality. Expressed in theoretical terms, there should be a sufficient number of tight links on the evaluator side to make the inscription device "solid" and "trustworthy", while the links that would otherwise support the evaluation object should be weakened or made "loose". Our interest in connections and disconnections should not be translated too quickly and directly into operational empirical analysis, since there is always a large number of connections and disconnections that can be studied, and since we can imagine a whole set of repertoires that can be relevant in the production of a contestability differential, ranging from norms, beliefs and resources to regulatory or coercive institutional mechanisms (Scott, 1995). Yet, the idea that it should be possible to empirically trace how a contestability differential might be produced remains metaphorically and conceptually promising. Statistical comparison, if regarded as trustworthy may be an important strategy here, because comparison always by definition leaves some unit with a lower score than others.
While it should be left to empirical analysis whether a given inscription device is in fact successful, the idea of the contestability differential leads to a productive analytical question: How might it be possible for a given inscription device to align itself with elements in a whole configuration of social forces so that it maximizes its contestability differential vis-a-vis the practices it describes, measures and evaluates? What are the links--and disconnections--we should look for to understand how global inscription devices become operationally possible?
In this light, the following questions represent more specific issues that a good theory of inscription devices must deal with:
* How does a global inscription device link to a particular institutional framework?
* How does a global inscription device produce comparable measurements?
* How does a global inscription device link to the media?
* How does a global inscription device enrol local actors?
* How is a global inscription device socially productive?
3. How Does a Global Inscription Device
Link to a Particular Institutional Framework?
An insight from the sociology of knowledge is that any production of knowledge is dependent on an institutional hinterland (Bijker and Law, 1992). For example, it is interesting to note that the OECD as the institutional hinterland of PISA has no command over legal nor financial resources that would put this organization in a position to exert direct influence over national educational policies (Grek, 2012, p. 24). Instead, the facilitation of comparative statistics in the auspices of OECD seems to be its most important instrument; it is more than a method, but should rather be regarded as a part of policy-making itself (Novoa, 2002, p. 144). Policy instruments are interpretive devices that shape the world and create a space in which policies can be carried out in particular ways. "Every public policy instrument constitutes a condensed and finalized form of knowledge about social control and ways of exercising it" (Lascoumes and Le Gales, 2007, p. 11).
The affiliation between OECD and PISA is also important because it supplies PISA with a particular institutional imprint (Scott, 1987). Evaluation devices tend to represent the values and world views of the institutions that produce them. PISA is thus imprinted with a view of the world as primarily economic and the relation between its elements as primarily competitive. The problematizations of the world that OECD takes for granted are those problematizations to which PISA presents itself as a management tool. In a PISA perspective, educational statistics are keys to understanding how educational policies produce results that prepare national populations for international economic competition.
While educational researchers may argue that the link between PISA results and future economic competitiveness may be theoretically and empirically weak, this observation, although technically correct, does not constitute sufficient grounds for claiming that the appeal to economic competitiveness is purely ideological (in the classical Marxist sense of a "false belief'). If politicians believe that continued GDP growth is important to maintain social order and if populations are sufficiently concerned about unemployment, crisis, social dumping, global competition and declining welfare services, then any promise that claims to prepare a country for an uncertain economic future may exploit these fears and anxieties. Imaginations of an uncertain economic future are real enough, and if a global inscription device such as PISA symbolically enrols these imaginations, understanding the link between imaginations of economic futures and educational policies may help us understand how PISA works.
While being imprinted with a world view representing the ideals of economic growth, PISA links itself to other dominant beliefs and institutions in contemporary Western society. It presents itself as a tool that helps ameliorate economic problems that are already on the agenda in modern societies while at the same time highlighting only one aspect of education, i.e. that which has to do with economic growth. In that sense, a link to an institutional framework is also a link to a particular world view (Berger and Luckmann, 1966), in the case of PISA an economic one.
However, a link to an institutional framework is more than that. It is also a link to a particular governance structure such as a set of hierarchical relations and patterns of accountability. Although, of course, OECD consists of member states, and although each country decides for itself whether it wishes to participate in PISA, the definition of success criteria in PISA is hardly an open democratic process. The very task of producing comparative information presupposes a decoupling of PISA criteria from specific national goals in education that would otherwise in a classical national perspective be seen as legitimate products of a legitimate democratic process. Moving an inscription device from a national to an international arena is equivalent to moving it into more uncertain and unclear democratic territory.
If there is a democratic deficit in the present international political order, an interesting question is thus which institutional mechanisms can be used for stakeholders who are affected by a global inscription device but who have no role in its official institutional governance structure. In the case of credit ratings, we have seen legal cases in which attempts are made to hold credit rating agencies accountable for valuations that turned out to be disastrous for those who trusted them.
So, even if a given inscription device is not part of a conventional political accountability structure, there is no such thing as a "private" or "independent" production of knowledge in abstract generality. The question is which institutional mechanisms are available or can be made available so that the overflow of social effects produced by an inscription device can be re-circulated back to the producer of the device.
Although today not enough is known about the governance mechanisms and institutional channels of influence to which global inscription devices must respond, the issue continues to be interesting because their operations and their inscriptions are influenced by the these institutional features. The question is whether new institutional orders can be created that increase the sensibility of producers of global inscription towards the effects they create.
Among the institutional mechanisms that should be taken into account are the self-sustaining processes that may help strengthen a particular institution responsible for a global inscription device. If the inscription process requires expertise, then critique of the process induces the institution to hire more expertise. The resources necessary to produce global inscriptions in the style of PISA are huge, and the higher the number of countries that join PISA, the higher are the costs of potentially producing an alternative to PISA, because one of the strengths of PISA is exactly the comparative possibilities in a very large data set. After some time, it becomes practically impossible to even think of an alternative.
Nevertheless, the self-sustaining processes supporting an institution that produces a global inscription device may involve a delicate and precarious balancing act between various connections and disconnections. In PISA, for example, an extensive effort was made to enroll expertise, including researchers, in the construction and implementation of PISA surveys at both international and national levels. This move may have helped increase the legitimacy of PISA in some circles. It is a common norm among researchers that all methodological information be disclosed in order to allow critical discussion of the chain of evidence that leads to results. In PISA, however, most of the actual tests are kept secret from one year to another so that no pupils can learn automatic responses and thereby undermine the validity of the tests. At the same time, however, this principle violates the scientific norm of full disclosure. As a consequence, PISA may lose legitimacy in some circles, and the participating researchers may suffer from their double role. In other words, the institutional links that promote PISA in one sense may create disconnections in another, so the institutional balancing acts in making a global inscription device operate may be more precarious than we would have thought at first sight.
4. How Does a Global Inscription Device Produce Comparable Measurements?
A statistical tool does not only measure something that already exists. First it is necessary to make up those categories on which measurement can be based. These include individual and collective identities to which numbers can be ascribed (Porter, 1995, p. 42) as well as attributes, qualities, properties, performances and other characteristics under measurement. Categories are often loosely defined until the process of measurement requires an exact operational definition. Quite a few phenomena are vague, interpretable social constructions that only become defined more specifically because they need an operational definition to become measurable. Phenomena such as intelligence, achievement and performance are often primarily understood through particular technologies of measurement. A description device that requires statistical operations therefore has the capacity to help define and constitute the phenomenon it claims to measure.
Educational statistics are sometimes faced with some very practical operational problems that in fact point to larger conceptual and political issues. Even a simple definition of "a country" may be complicated faced with a decision about whether, say, Belgium is one country or whether it should be described in terms of its Flemish and Wallonian parts, respectively. If pupils don't start school at the same age in different countries, should PISA compare pupils at a particular age or pupils in a particular grade? How are forms of education that are unique to a particular country represented in comparative statistics? Quite a lot of operational work may be necessary for a particular inscription device to actually produce inscriptions because the different national contexts in which it operates may be rough and difficult to compare, but it is the function of these operations to smoothen out the differences so that comparable inscriptions can be made.
In order to produce global comparisons of the performance of national education systems, it is necessary to produce a new set of common criteria. PISA criteria are a product of a mix of concerns for "lifelong learning", "cross-curricular competencies", "learning strategies", and the desire to support "policy lessons" (Grek, 2009, p. 27). In inventing a set of self-devised evaluation criteria, PISA does several things at the same time. It decouples itself from national policy goals, as the use of these goals as standards would not facilitate comparison. PISA also "makes itself necessary": While PISA creates a new and common imaginary space in which policy-making can take place, only PISA is in a position to provide the data that describe how each country is located in the same statistical universe.
The perhaps most fundamental question in comparative statistics is whether it is possible to make sure that a phenomenon under measurement actually means the same thing to people in different countries. It is appropriate here to remind native English speakers that terms equivalent to "achievement" and "performance" simply do not exist in many other languages. On the other hand, other languages also hold certain key terms, such as "Bildung," central to educational thinking, while this term may not be directly translatable into English. If we respect the deep, historical, contextual and situated productions of meaning in education (Novoa and Yariv-Mashal, 2003), it is impossible to believe that the establishment of any comparative, statistical language (whether or not it happens to be based in English) takes place without considerable costs. However, loss of meaning is perspectivedependent. In order to compensate for loss of meaning, producers of one meaning system may listen to other meaning systems in order to find out what can be translated. In doing so, however, if one meaning system (A) is in position to demand that another meaning system (B) presents its statements in such a way that they can be understood from A's perspective, that is, although well-intended, equivalent of establishing power over B because B is not in position to demand the same thing from A (Becker, 1996). Perhaps the question of above--whether indicators mean the same to people in different national contexts--was incorrectly formulated. Perhaps it does not matter that specific meanings from national and local contexts are ignored or glossed over; what matters is that another transportable meaning system is produced from which these specific differences do not count if they cannot be translated into a global vocabulary. This functional superiority of a global inscription device may be more important than its perceived lack of methodological validity.
Indeed, one of the main accomplishments of a successful global inscription device is that it produces a common set of categories, a joint imagined educational space within which educational policies and activities can take place (Novoa and Yariv-Mashal, 2003; Ozga et al., 2011). From a discursive perspective, it is a great advantage that a vocabulary is established that makes it possible to talk effectively about educational issues in such common space. It may be the very constitution of such a common language that constitutes success for a global inscription device.
If that is true, then a critique of a description device for having low validity (meaning: it does not describe correctly the reality it claims to measure) is based on an inadequate understanding of the performative and constitutive capacities of such a device. To say that it incorrectly represents reality may be to miss how it, at least partially, defines anew what it claims to measure (Dahler-Larsen, forthcoming).
5. How Does a Global Inscription Device Link to the Media?
To inscribe is to leave visible traces. Without a link to mediascapes, global inscription devices may be invisible and thus deprived of socially constructive power. A discursive mise-en-scene of a spectacle undergirded by educational statistics is a forceful tool in education (Stronach, 1999).
Some aspects of inscription devices are particularly useful in relation to the production of spectacular news. Numbers have an air of facticity. Over time, the ranking of any particular country goes up and down. Whether or not a specific change in a position in a league table is a result of an increased number of countries participating, and whether or not it is within the limits of statistical uncertainty, media often report stories on whether a country goes up or down in relative ranking. Comparative numbers set the stage for stories about winners and losers in a competitive game.
These imaginative narratives can be linked to stories about heroes and villains that are held responsible for the results of national policies in education. In these ways, statistical inscription devices easily lend themselves to the type of dramatizations that circulate in media. However, it is not the function of modern media to produce either wisdom or consensus, but rather to produce what Luhmann (2002) calls "irritations", i.e. endless series of "problems" with "solutions" that lead to new "problems" (Luhmann, 2002, p. 100).
Therefore, various actors can engage themselves in the debate over the meanings and implications of global inscription devices such as PISA. A minister of education may decide not to let the exact ranking of a country be mentioned in a press release. He or she perhaps seeks to establish a particular framing of PISA data in the situation. The omission of ranking numbers may or may not become a news story in itself.
Critical researchers may find space in media to claim that PISA figures are unreliable or used in ways that cannot be justified scientifically. Various initiatives already planned may find new justification when they are presented as if they are a response to PISA results. A politician that reaps political benefits from criticizing a government based on poor PISA scores may later become a minister of education and then run into trouble if PISA figures have not improved. In other words, although there is an affinity between the logics of a global inscription device like PISA and the logics of media, the continued production of new news in non-linear ways should not be underestimated. However, this observation may only help to underline the continued socially constructive character of inscription devices reported in media. In other words, inscription devices such as rankings and tests on the one hand and media on the other stand in a mutually dependent relation with each other, but one that must be dynamic to be productive.
6. How Does a Global Inscription Device Enrol Local Actors?
The very production of actorhood is a key theme in many variations of social constructionism. In that respect, global competitive structures are interesting social constructions. From the view of each participant, the existence of the other competitors is a fact that incites each actor to define his or her actorship and start behaving in rational and controlled ways in order to win as good a position in the competition as possible.
Global inscription devices help produce actorhood by ascribing characteristics and scores to subjects and by defining them through their embeddedness in comparative and competitive relations. Nations become defined as global competitors. Actors need to make policies for nations. It is thus too simplistic to analyse educational policies in terms of what goes on at the global level and what goes on at the national level, respectively. Instead, education should be regarded simultaneously through a global and a national eye (Novoa and Yariv-Mashal, 2003, p. 425). The role of a global inscription device may actually be to set the stage for national actorhood in educational policy-making. Together with issues of security, education remains one of the policy areas in which the nation state still wants to exercise visible political power.
One of the forces of a global inscription device like PISA may precisely rest in its ability to enrol national policy makers as visible and active agents in the making of educational policy. While some institutional theorists suggests that imitation is a key mechanism that creates uniformity and isomorphism in a whole institutional field (here the global educational policy space) (DiMaggio and Powell, 1983), Meyer (2010, p. 13) argues that the imitation metaphor may be ill chosen because political actors are often highly conscious of their rationales and their scripted motives. The more highly stylized the policy, the easier it is to articulate such rationales. A global inscription device may help provide the script and the vocabulary that a national actor needs in order to explicate and rationalize a stylized policy initiative (Meyer, 2010, p. 13).
The mobilization of such rational and scripted actorship may be more, not less, theoretically relevant if actors play this role in partly contextualized and partly improvised ways. For example, Grek, Lawn and Ozga (2009) showed how Scottish policy makers used PISA to justify their own internal quality assurance processes and promote Scotland's separate and distinctive education system. The fact that some level of interpretive flexibility is possible around a social construction so that it mobilises diverse actors (Pinch and Bijker, 1984) may in fact increase its use and its circulation rather than opposite.
As Meyer (2010, p. 13) explains, however, socially constructed actors are in demand far beyond realistic expectations. A national environment of educational policy-making is likely to be so full of conflicting political and practical pressures that it is often not possible to directly translate a pure recommendation based on any global inscription device directly into policymaking. Expectations are so many that even socially constructed actors cannot do everything that is expected of them.
Institutional theory therefore predicts very loose couplings between measurements and initiatives, strategies and actions, intentions and results. These loose couplings exist not only because national contexts for policymaking are complex and multi-dimensional, but also because it is not always easy to build a logical step from a PISA-score to a policy-making initiative (although OECD and others of course help devise policy initiatives, too). An even more fundamental impossibility is built into international competitive comparison. Positions in comparisons are, by definition, positional goods that not everyone can have. Not every nation can be a winner. A logical chain of events from PISA results to initiatives designed to produce PISA success must therefore be broken somehow somewhere most of the time for most actors.
It is therefore inadequate to criticize global inscription devices such as PISA for lending itself to half-cooked, inconsistent and basically unjustified policy initiatives which take place in the name of PISA. Advocates of this type of critique would probably not be satisfied anyway if PISA really did lead consistently to only one set of educational practices.
The theoretical point that should not be missed is the double nature of mobilization of a supposedly rational national policy maker that acts explicitly, consciously and creatively in more or less stylized way in a strategic imaginary landscape the dimensions of which are defined by global inscription devices. At the same time, contemporary observations consistent with "The Affective Turn" in social theory (Clough, 2007; Cole, 2011) also suggest that inscription devices operate through anxiousness, nervousness and fear (Lingard and Sellar, 2013, p. 639).
7. How Is a Global Inscription Device Socially Productive?
The socially constructive character of global inscription devices cannot be fully appreciated unless we cultivate an intellectual curiosity about the many forms of its consequent productions. Responding to a conceptual need to describe these social constructions as more than merely unintended effects (Dahler-Larsen, forthcoming), concepts such as "catalyst data" data that "pressure politicians, policy makers and systems to respond" (Lingard and Sellar, 2013, p. 652) and "database effects" (Lingard and Sellar, 2013) or "constitutive effects" (Dahler-Larsen, forthcoming) are being developed. In the same spirit, and close to ANT, the concept of overflow (Callon, 1998; Cole, 2010) can be applied in a nuanced understanding of how various actors at different levels respond to the consequences of global inscription devices in their own way and thus create new practices.
We have already pointed out the definition of a particular discursive space for educational policy-making and the appeal to rational national actors to compete within that space. Both of these phenomena have social overflows, i.e. they lead to additional sets of social constructions, new links, new connections, new justifications. For example, the rational national policy-maker described above is an abstraction politically speaking. Policymaking involves several actors with various interests. The distinction between government and opposition also plays an important role in attributing blame for bad results and taking credit for good ones. If you are in opposition, the ambiguity of potentially taking over government one day and not having the time or the power to fundamentally change PISA scores introduces an element of uncertainty into one's critique unless, of course, the memories of voters and journalists are so short that the only political points that count are those scored today.
It is true, however, that national debate over educational policy is more conflictual and contested than one would expect if an international inscription device such as PISA totally defined the policy space as such. It may happen, however, that a majority of political interests regard improved PISA scores as an accepted goal in educational policy and only disagree about the means to get there, and that is in itself a half victory for a global inscription device. The establishment of joint taken-for-granted goals is an important milestone in a socially constructive process, even if the choice between specific initiatives remains. One of the reasons why shared goals are important is that they help create a distinction between legitimate interests who accept that goal and peripheral interests who do not and thus cannot take part in policy-making processes.
Among the overflow effects is also the export of new evaluative inscription devices at various levels of the education system. If the position of a nation in an international comparison is contingent and therefore infused with risk, the nation may respond by exerting more control over units internally in the country which help produce national outcomes. In this light, PISA may help justify a whole range of tests, comparisons, statistics, quality reporting mechanisms, accreditation and auditing procedures at the levels of municipalities, schools, teachers and pupils. These new inscription devices, closer to school principals and teachers, may produce a whole new set of constitutive effects (Dahler-Larsen, forthcoming) and overflows.
Evaluation tools at the local level may help stabilize a productive mentality that is aligned with the mentality inherent in a global inscription device, but may also induce new risks at local levels (Rothstein, Huber and Gaskell, 2006). For example, schools in socioeconomically and ethnically underprivileged areas will object to being compared directly to schools in other areas, but if an improved statistical analysis is provided that controls for socioeconomic background variables such as the number of single mothers in a school district, then the single mothers may object to publishing that control variable.
Sometimes, the responsibility for crafting new inscription devices in detail is pushed down to those levels where they are supposed to be implemented. Such strategies may be presented as participatory and democratic initiatives with the aim of enhancing the relevance, meaningfulness and usability of these evaluative tools among local actors. Difficulties sometimes arise, however, for those who are involved in defining the operational characteristics of new inscription devices, because it is difficult to measure educational outcomes in ways that are acceptable to local educators and teachers. When the construction of inscription devices moves towards the local level, the anonymity of their constructor is broken, and critique may be directed more directly to him or her compared to what we see in the case of large-scale anonymous evaluation machines (Dahler-Larsen, 2012).
In other words, risk that is exported from the national level to other levels in the political system does not disappear; instead new actors that become mobilized push it back or seek to promote risk-reducing initiatives such as the allocation of additional funding to under-privileged schools, perhaps leading to spiralling dislocations of risk (Rothstein, Huber and Gaskell, 2006). These examples suffice to show that the productive character of educational inscription devices can best be understood as open-ended. The consequent socially constructive processes spread across time and place.
If these effects of global inscription devices take place at social arenas that are perceived to be separate in time and place from a given global inscription device, i.e. if the simultaneousness and interconnectedness of the "local" and the "global" is not attended to, then it becomes difficult to define and determine the many effects flowing from global inscription devices. Perhaps one of the main productive strengths is that their effects appear below the radar of our conventional distinctions such as global/local, political/non-political, intended/unintended etc.
Global inscription devices operate through links that partly escape or circumvent conventional national and democratic systems for legitimate decision-making, but exercise power and create politics through a number of surprising links and connections in various directions at various levels. They may also work because once enfolded in the inscription device, central assumptions are black-boxed and disappear from the radar screen.
A Case Example: The Slowest Students in the World While I was writing this article, the newly appointed minister of education in Denmark, Sofie Carsten Nielsen, talked about the need to reform higher education because, as she said, "the Danish students are the slowest students in the world." A radio program, "Detektor", that looks critically at how data are used by authorities in the media, asked the ministry to deliver data in support of that statement and was given a short note that referred to excerpts from Education at a Glance (OECD, 2013). As a professor in evaluation, I was asked to assess whether the data in fact did lend support to the minister's statement. During a lengthy interview (not on the air), I explained three things. First, the minister used data about age at graduation to make conclusions about length of study, thereby confusing the two (OECD claims to describe the former only, not the latter). Next, the minister focused on age of graduation with a Bachelor's degree rather than with a Master's degree, although most professions in Denmark require a full Master's degree. I therefore found the latter more relevant, but on that indicator Denmark had a favourable score, better than OECD average. Finally, I asked the question why students need to complete their studies so quickly? Because, according to OECD, the only way we can know the economic value of employees is through their productivity (for which salary is used as a proxy variable in the case of public employees). In the OECD area in toto, there is a strong economic effect of completing an education because of the wage differential between people with higher education and those without it. However, in the egalitarian Danish society there is only a small wage differential. In other words, an economic calculation that is valid in OECD in general but is less valid in Denmark is tacitly undergirding the justification to speed up higher education programs also in Denmark.
In the final radio program (which aired on March 6, 2014), two succinct passages relating to my two first points were included. The minister replied that she was sorry she had overstated her point, and she promised to be more careful with data in the future, although she still said the need for reform is pressing.
Relating this case example to my five questions, we clearly see, of course, the assumptions of OECD as an economic institution providing the very epistemic skeleton of the datascape that is referred to. Education at a Glance even attributes itself a degree of institutional autonomy (from national democratic governments and from OECD itself!) by stating that "This work is published on the responsibility of the Secretary-General of the OECD. The opinions expressed and arguments employed herein do not necessarily reflect the official views of the Organisation or of the governments of its member countries."
Education at a Glance makes itself available as media news as the elements of speed, ranking and international competition provide an apparently attractive imagery. Paradoxically, this inscription device made itself relevant by providing facts that enrolled a newly appointed minister and made her withdraw statements that she had made too quickly. The falsification and the subsequent apology were also news. While she was firm in pursuing her reforms, she promised to be more careful with data in the future--perhaps more disciplined vis-a-vis the datascapes she can only use, but not control?
It remains to be asked why my third and more controversial point about wage differentials and the (withering) justification for speeding up higher education was not cited in the radio program. Perhaps it was too lengthy and difficult to explain. Perhaps it ran counter to accepted economic ideology. Perhaps the radio program was victim of its own pre-cooked narrative dramaturgy: See how quickly we can catch the minister in making a mistake.
Reflecting on the story, I am sure that the media, Education at a Glance, the minister, and myself interacted for a short while, but I am not sure who were being used by whom. At the end of the day, however, certain statements were withdrawn, and certain "facts" established, to which the minister promised to be loyal. A fundamental problematic assumption, however, was never questioned. In this case, a contestability differential was kept in place by keeping that very assumption away from the radar screen.
8. Conclusion and Perspectives
Inscription devices are instruments that aid comprehension typically but not solely by rendering objects or a phenomenon visible and comparable. In addition, global scales operate with quantification that implies its own interpretations and definitions, including an abstraction from qualitative differences in what is being measured.
A global inscription device co-produces a particular understanding of educational goals assuming that it is relevant to compare countries along the same quantitative dimensions since they share the same educational goals.
Although no full-fledged theory of global inscription devices exists, we have suggested that a social constructivist perspective that attends to links and connections as well as disconnections and loose couplings may be fruitful. We have also suggested a handful of more specific nodes and connection points that a good theory of global inscription devices should attend to. For example, the institutional framework in which a global inscription device is embedded may provide a set of resources and a world view that supports a particular type of inscription. At the same time, such an institutional framework also channels and filters various types of regulation, feedback and potential critique to which a global inscriptive practice is exposed.
We have also suggested that the link between an inscription device and the modern mediascape is crucial. Furthermore, global inscription devices work through the mobilization of national political actors that so to speak step into the strategic landscape defined by comparative measurements. Last, but not least, we have proposed that a global inscription device may be highly productive for example through the indirect instalment of a number of evaluative tools and supra-national levels. A national system of evaluation machines may be necessary not because of its effects upon education (see Cole, 2012), but because it supports a national actor in policy making in reporting back on how he or she responds to the challenges produced by international comparisons.
One of the forceful ways in which a global inscription device operates may exactly be through these dynamic and multifaceted processes that create links across and beyond many of our classical conceptual dichotomies such as global/local, political/unpolitical, and intended/unintended.
In other words, global inscription devices may operate effectively partly because they create new links to relevant sets of institutions, beliefs, media, and actors, and partly because they circumvent some of our conventional ways of democratic debate, resistance, regulation and perhaps hesitation. Global inscription devices establish a contestability differential that makes them operational and functional both through connections and through disconnections.
In this light, the effects of global inscription devices upon educational policy and practice may be far-reaching and of considerable magnitude, while often not totally acknowledged, nor understood. Although our analysis has demonstrated the analytic value of a handful of specific how-questions that
help us understand the stabilization of a contestability differential in and around a global inscription device, it remains an interesting and challenging dilemma to explain the effects of such global inscription devices while respecting the vulnerability and fragility of the many operations through which they are constructed.
According to the approach advocated in this article the effects of global inscription devices deserve to be studied rather than just assumed, since they depend on the construction of a complex set of links and connections as well as disconnections. Our insistence that institutional arrangements are key to understanding the production of global inscriptions raise the question of whether innovative institutional arrangements would pave the way for new kinds of global inscription devices that would be more innovative, more democratic and would appeal to alternative imaginations in contrast to those we see today.
Appadurai, A. (1996), Modernity at Large. Cultural Dimensions of Globalization. Minneapolis, MN: University of Minnesota Press.
Ball, S. (2009), "Academies in Context: Politics, Business and Philanthropy", Management in Education 23(3): 100-103.
Becker, H. S. (1996), "The Epistemology of Qualitative Research", in R. Jessor, A. Colby, and R. Schweder (eds.), Ethnography and Human Development: Context and Meaning in Social Inquiry. Chicago, IL: University of Chicago Press, 101123.
Berger P., and Luckmann, T. (1966), The Social Construction of Reality. Garden City, NY: Doubleday.
Berger, P., Berger B., and Kellner, H. (1973), The Homeless Mind. Oxford: Penguin.
Bijker, W. E., and Law, J. (eds.) (1992), Shaping Technology/Building Society. Cambridge, MA: Massachusetts.
Callon, M. (1998), "An Essay of Framing and Overflowing: Economic Externalities Revisited by Sociology", in M. Callon, (ed.), The Laws of the Market. Oxford: Blackwell, 1-57.
Clough, P. (ed.) (2007), The Affective Turn. Durham, NC: Duke University Press.
Cohen, I. B. (2005), The Triumph of Numbers. New York: Norton.
Cole, D.R. (2010), "The Reproduction of Philosophical Bodies in Education with Language," Educational Philosophy and Theory 42(8): 816-829.
Cole, D.R. (2011), "The Actions of Affect in Deleuze--Others Using Language and the Language that We Make ..." Educational Philosophy and Theory 43(6): 549-561.
Cole, D.R. (ed.) (2012), Surviving Economic Crises through Education. New York: Peter Lang.
Dahler-Larsen, P. (2012): The Evaluation Society. Palo Alto, CA: Stanford University Press.
Dahler-Larsen, P. (forthcoming), "Constitutive Effects of Performance Indicators. Getting beyond Unintended Consequences", Public Management Review 16.
Grek, S. (2009), "Governing by Numbers: the PISA 'effect' in Europe", Journal of Education Policy 24(1): 23-37.
Grek, S., M. Lawn, and Ozga, J. (2009), "PISA and the Policy Debate in Scotland: Policy Narratives about Scottish Participation in the International Comparison", Sisifo. Educational Sciences Journal 10: 73-84.
Koselleck, R. (2006), "Crisis", Journal of the History of Ideas 67(2): 357-400.
Lascoumes, P., and le Gales, P. (2007), "Introduction: Understanding Public Policy through Its Instruments--From the Nature of Instruments to the Sociology of Public Policy Instrumentation", Governance: An International Journal of Policy, Administration, and Institutions 20(1): 1-21.
Latour, B. (2005), Reassembling the Social. An Introduction to Actor-NetworkTheory. Oxford: Oxford University Press.
Lingard, B., and Sellar, S. (2013), "'Catalyst Data': Perverse Systemic Effects of Audit and Accountability in Australian Schooling," Journal of Education Policy 28(5): 634-656.
Luhmann, N. (1996), Die Realitat der Massenmedien. Opladen: Westdeutscher Verlag.
Mark, M, G. Henry, and Julnes, G. (2000), Evaluation. San Francisco, CA: Jossey Bass.
Meyer, J. W. (2008), "Building Education for a World Society", in M. A. Pereyra (ed.), Changing Knowledge and Education. Frankfurt: Peter Lang, 31-50.
Meyer, J. W. (2012), "World Society, Institutional Theories, and the Actor," Annual Review of Sociology 36(1): 1-20.
Ozga, J., P. Dahler-Larsen, C. Segerholm, and Simola, H. (2011), Fabricating Quality in Education. London: Routledge.
Pinch, T.J., and Bijker, W.E. (1984), "The Social Construction of Facts and Artifacts: or How the Sociology of Science and the Sociology of Technology might Benefit Each Other," Social Studies of Science 14(3): 399-441.
Porter, T. (1995), Trust in Numbers. Princeton, NJ: Princeton University Press.
Rothstein, H., M. Huber, and Gaskell, G. (2006), "A Theory of Risk Colonization: the Spiraling Regulatory Logics of Societal and Institutional Risk", Economy and Society 35 (1): 91-112.
Scott, W. R. (1995), Institutions and Organizations. Thousand Oaks, CA: Sage.
Stehr, N. (2001), The Fragility of Modern Societies. London: Sage.
Stronach, I. (1999), "Shouting Theatre in a Crowded Fire: Educational Effectiveness as Cultural Performance", Evaluation 5(2): 173-193.
Thompson, G., and Cook, I. (2014), "Becoming-topologies of Education: Deformations, Networks and the Database Effect", Discourse: Studies in the Cultural Politics of Education. Published online 26 feb 2014.
Department of Political Science, University of Copenhagen
Peter Dahler-Larsen is professor of public administration at the Department of Political Science, University of Copenhagen. His research interests focus on institutional, cultural and political aspects of evaluation systems. He is past president of European Evaluation Society, and author of The Evaluation Society (Stanford University Press, 2012).
|Printer friendly Cite/link Email Feedback|
|Date:||Jul 1, 2014|
|Previous Article:||International students and the politics of language among 'Globalising Universities' in Asia.|
|Next Article:||Higher education development and knowledge economy optimism in Ethiopia.|