1 Tribute to those who saw the Birnam Wood: the Italian School of Phenomenological Pedagogy

At the end of the ‘80s and throughout the ’90s of the XX^th century, Piero Bertolini (1931–2006) and a group of Italian scholars who drove the Italian School of Phenomenological Pedagogy, focused on an issue that — at the time — appeared relatively irrelevant, at least with respect to mainstream research on education. What they pointed to was the loss of the sense of education face to the advancing of a specific ’sense of education’, the one inscribed in the parametric hypertrophy, the evaluation pressure, and the “accountability totem” (or “régime,” Biesta, 2004; see Caronia, 2022) that were increasingly acquiring credibility and even enthusiastic support. Three works have been the pillars of such an unfortunately largely unheard voice: Postprogrammazione [Postprogramming], by Gabriele Boselli (1998); Sulla didattica [On Didactics], edited by Piero Bertolini (1999a) and La valutazione possibile [The Possible Assessment], edited by Piero Bertolini (1999b). As a Cassandra’s warning, these works — where scholars in pedagogy, philosophers of education and school professionals intertwined their voices — did not make a difference in the Italian school system but certainly largely anticipated actual concerns about some — perverse yet expectable — outcomes of the parametric hypertrophy and the evaluation pressure turns in education (see below). Boselli (1998) and Bertolini (1999a) pointed to the epistemic gerrymandering and knowledge segmentation that were the prodrome to the construction of measurable “didactic units” or “learning objects” constructed as assessable and verifiable items. This deliberate construction of what a “learnable” should look like was nothing but the epistemic bases of what — a couple of decades later — would be the evidence-based education turn. The third volume (Bertolini, 1999b) was an upstream masterpiece in the Italian landscape of works dealing with assessment in education. It directed the audience’s attention toward the explosive density of a principle now known as “accountability.” Introduced gradually and quite surreptitiously in the school system, it turned out to be a kind of Trojan horse: coupled with the measurement and assessment turn, it produced the logical reversal between the sense of education and the forms of its evaluation.

Following the legacy of these scholars and building upon their work and vision, I advance that in the era of school autonomy, accountability is not only an understandable practice but even a necessary one: public schools should be accountable to the stakeholders, and their decisions, policies and practices should be submitted to transparency, inspectionability and comparability. As I will try to demonstrate in this article, the very issue at stake is not “accountability” as such, but the “vocabulary of motives” (Wright Mills, 1940) that orients the pursuit of social acceptability. I will come back to this point in the conclusion. After a very brief premise on the actual umpteenth return to objectivism, I will turn to a) the invisible solidarity between the “evidence-based practice” claim and the accountability turn in school and care institutions and b) the haunting question of values in the era of evidence-based educational decision-making (EBEDM). Then, I will delineate three risks (beyond the already identified “learning to test” and “learnification” risks: Biesta, 2004, 2010) implied in pursuing an objectivistic foundation and — increasingly — an economist ratio of educational policies and practices. Before concluding by claiming for “moral accountability,” I will delineate the reasons and consequences of, and alternatives to, what I call the “de-moralization” of educational decision-making (Caronia, 2022a). 

2 The evidence-base turn, or the “what works” cargo cult¹

Despite decades of epistemological warnings on the social construction of scientific knowledge and the tribute it pays to the conditions of its production, we are witnessing a (renewed) paradigmatic consensus on empiricism in (social) sciences and a shared trust in evidence-based practices and decision-making. I do not argue against the necessity of empirical accuracy, nor do I question the suitability of measuring and assessing some dimensions/features of educational practices. Rather I focus on the educational decision-making in the era of:

1. evidence-based practices;

2. efficacy and efficiency principles;

3. accountability.

Many fields of study tend to produce what Bourdieu called “the naturalization of their own arbitrariness” (Bourdieu, 1977, p. 164). However, not only are some of them particularly committed to producing “evidences” for implementing social policies and practices, but they are expected to do so: health care, education, psychology and nursing are, understandably, among them. The question of course concerns what is considered to be evidence. A hierarchy of evidences (and therefore a hierarchy of the different research methods producing them) has already been established and slowly migrated from evidence-based decision-making in medicine (Cochrane, 1972; Evidence-Based Medicine Working Group, 1992) to evidence-based policies and practices in education and other applied sciences (for a plea in favor of evidence-based education see among others Slavin, 2002, 2004; for a critical stance see Howe, 2004). Basically this approach (for some even a paradigm) establishes that decisions should be based on evidence, evidence should originate from empirical research, and the best evidences are those that come from randomized controlled trials or analogous scientific research designs giving an epistemological primacy to quantitative-experimental studies. Although the evidence-based approach has been deeply criticized even within medical studies (for an overview of the criticisms to evidence-based medicine see Cohen, Stavri, & Hersh, 2004), it still has a strong appeal.

Nothing more than the recent health crisis has displayed our public faith in the “expert,” its being systematically convoked and invoked as a new oracle, the “subject supposed to know” how things are going and, moreover, how things will go. The voice of science, embodied in the expert, “animates” decision-making that appears and should appear as nothing less and nothing more than the operationalization of the expert’s voice. Policy makers and decision makers should listen to this voice and translate it in practical terms as if good, justifiable governance should be a value-free, non-discretional corollary of scientific knowledge. Not surprisingly, the subsequent step of such a path from “evidence to action” is the increasingly delivered “guidelines” or protocols, i.e. textual inscriptions of operator-free praxis that should appear as if it was a corollary of scientific evidence and not the outcome of ethical, political or even local (educational) arguments. Ideologies, values, moral horizons, cultural models, situated phronesis and whatever criteria other than “scientific evidence” has no right to be invoked as a drive, at least de iure. Indeed, and interestingly enough, these non-epistemic dimensions are de facto embedded in educational research practices that produce the “evidence” used to justify policies (for an illustration see Caronia, 2018, 2022b; Caron & Caronia, 2019). Furthermore, not only do non-epistemic values lead scientific research, but they inhabit educational policies and practices far more than the evidence-based approach expects or even wishes. What are the risks of concealing the value-ladeness of (research in) education under the disguised objectivism of the evidence–based turn in education? Why do we so obdurately pursue such a (seemingly) value-free dimension of educational decision-making? After discussing the risks implied in pursuing a de-moralized decision-making in education, in the next sections I advance some answers to these questions.

3 The risks of obdurate objectivism and economistic turn in education

Adding to the “teach to test” and the “learnification” (Biesta, 2010) risks, I contend that contemporary EBDM coupled with the economist turn in education runs three further risks. The first risk, or the realistic fallacy, consists in ignoring the mundane — and therefore cultural — roots of “scientific evidence” mobilized as objective bases of educational decision-making and practices. As sociologists and philosophers of sciences demonstrated long ago, the objectivism of evidence is a meaning effect of identifiable specific procedures and rhetorical resources through which we textualize scientific findings as well their modes of production (for an account see Caronia, 2018; Caron & Caronia, 2019; Caronia, 2022b).

The second risk is what I call the reductionistic fallacy. It proceeds from the “tunnel vision bias” implied in pursuing objectivity and the efficacy/efficiency economistic ratio. It consists in ignoring the relevance of what escapes a gaze oriented to taking into account only what can be segmented, measured according to a same and consistent unit of measurement, and assessed on a scale. A paramount example of such unmeasurable yet operating dimensions of education is the moral order inscribed in the discursive practices and the language of education (see Bruner, 1986; Ochs, 2002; Ochs & Schieffelin, 1993, 2012; for an illustration see Caronia, 2021).

I call the third risk the perspective fallacy. It is another consequence of the tunnel vision bias and consists in ignoring the (relative) irrelevance of the EB principle for most decisions we ordinarily take when engaged in educational decision-making. Do we really need evidence to claim for the “goodness” of letting children with cognitive impairments attend ordinary school? And if so, which kind of evidence would prove that this decision is better than the opposite, i.e. the “special schools for special educational needs” policy? It is not unthinkable that it would be possible to “provide” evidence sustaining each decision (e.g. the construction of a more inclusive mindset in pupils or a habit to interact with disabled people, vs. the achievement of better learning outcomes in inference making, logic, math or other curricular subjects), but the choice of which evidence we decide to produce and rely on is not — per se — a matter of evidence. We have known since Hume at least that descriptive statements cannot establish normative claims.

These three risks share a common premise — the pursuit of (seemingly) de-moralized educational decision making — and two consequences: first, a surreptitious and therefore invisible infiltration of the axiological dimension in educational policies and practices and, second, the neutralization of personal responsibility implied in undertaking any action, even when the action consists in following protocols, guidelines, plans or “evidence” (on instruction and instructed action as mutually constitutive, and therefore on the individual agency, interpretation and local production of an action as an “instructed one,” see Garfinkel, 1967; Suchman, 1987). Pursuing an evidence-based and therefore a seemingly de-moralized educational decision amounts to ratifying the assumption of the third-person perspective (e.g. following evidence-based guidelines) to account for individual action, abandoning the first-person perspective that characterizes (or should characterize, see Arendt, 1964/2003) the ethics of ordinary and professional everyday life (cfr. Caronia, 2022a, 2022b).

4 An answer to the left pending question: Why? The haunting question of values in the era of EBEDM

Notwithstanding the above-mentioned risks and their consequences, the pressure toward EBEDM is continuously increasing. Why? I advance that the contemporary umpteenth objectivist turn in education is a consequence of the growing demand for socially inspectable, socially justifiable and socially acceptable policies and practices. In a word, it is an outcome of contemporary pressure for accountability. As I mentioned above, the demand itself is more than understandable: the use of public resources needs to be justified to the stakeholders. The very issue at stake is “how”; how should these resources be accounted for? What is the “vocabulary of motive” (Wright Mills, 1940) of contemporary accountability?

In the last thirty years or so, the Italian school system (and before it the healthcare system) was assumed to be thinkable, organizable and governable as if it were a company. Within this business mold, the format of accountability increasingly used to account for a school’s — and therefore its principal’s — performances, a teacher’s work, yet also a residential community’s outcomes, is the economistic “audit model.” The audit model and its outcomes, i.e. certifications and accreditation, need measurements, internal and external verifications and assessments.

I advance that the audit model and relative pressure for finding measurable and assessable items oriented collectivities toward what seems to be the only possible domains of consent: the “objective” evidence, the costs and benefits ratio and the gap between the achieved and the expected outcomes. Basing decision-making on this kind of data is a way of excluding from the repertoire of socially acceptable justifications what seems to be divisive per se: values, the “right vs. wrong” category, the moral order nurtured by beliefs and justifiable certainties, ideal horizons, and temporarily shared worlds of meaning (Rommetveit, 1977) where a collectivity establishes a socio-historical version of the world as it is and “ought to be.” Any moral vocabulary seemingly lost its appeal and even legitimacy and the iteratively used, and therefore ratified and normalized (Derrida, 1988; Garfinkel, 1967), “vocabulary of motives” (Wright Mills, 1940) that frame social acceptability of education and healthcare systems is (constructed as) value-free (Caronia, 2022a).

It is extremely easy to understand why public governance avoids rushing in where angels fear to tread: values are divisive and the more a society is heterogeneous and pluralistic, the more they risk being so. Talking about values produces irreducible stances even about the nature of values themselves: as we know quite well, philosophers as well as laypersons are roughly positioned around two incompatible, irreconcilable frameworks as to the very nature of values: the realistic essentialism and the socio-historical constructivism (see De Monticelli, 2021; Colicchi, 2021). But, and here is my point, we do not evade the haunting question of values by making them invisible and evoking scientific knowledge and the benefits/costs ratio as if they were value-free. They are not: morality pervades even the seemingly most neutral search for evidence, yet it does it invisibly.

Not only, as previously argued, is scientific knowledge deeply yet invisibly value-laden (see among others, Longino, 1990, 2002; Douglas, 2007, 2009; Carrier, 2013), but so are the procedures that produce an objective portrait of an educational system or practice from their very outset, i.e. since the process of determining and paying attention only to what can be segmented into measurable entities (itemization). Any time we decide what we will measure and assess, what is an indicator of what; any time we define what counts as cost and what as benefit; any time we decide which is the temporal interval within which an objective should be achieved to be counted as “an achieved objective”; all these times we are paving the totally cultural and therefore moral field for our research of objective data, evidence and consensual value-free justifications of educational policies and practices. Clearly enough, then, they are not value-free; they are loaded with hidden — and therefore — unaccountable values.

I suggest that pursuing objectivism and the evidence-based educational decision-making is a deception, another Baconian idolum that makes us see what we want to see: “what works” packages apparently technical and free from values as well as other forms of cultural arbitrariness; practices easily accountable because they appear to be based on non-conflictual arguments and reasons for which it is easy to find public consensus, such as efficacy and efficiency, ratings and rankings.

5 Toward moral accountability of educational decision-making: some concluding remarks

According to Bourdieu (1977), (neo/post) positivistic social sciences cannot avoid their unavoidable quota of arbitrariness despite being concurrently oriented to concealing it, to give the appearance of — as much as possible — a mirror of (social) nature, an observer-independent portrait of how things are or will be under certain controlled circumstances. It is quite puzzling that this antinomic structure characterizes (or better, should allegedly characterize) scientific knowledge concerning care and education. Perhaps more than other sites of human sociality, these two ontological regions can and should deploy, exhibit and make socially inspectable their “arbitrariness,” i.e. their unavoidable dependency on the Life-world from which they originate and to which they should return as “relevant knowledge” providers. As mentioned above (see the perspective fallacy), do we need scientific evidence to decide that it is good and right to include cognitive disabled children in ordinary classrooms? If so, which kind of evidence will “prove” that this choice is better than its contrary? How can we found moral normativity on empirical evidence? And, moreover, why should we do? Why do we fear displaying the systems of values and conceal them under the ‘evidence-based education’ mantra?

In the previous section, I provided one possible answer to that question (but see also Caronia, 2022a): values are divisive while allegedly value-free scientific evidence and costs/benefits ratio gain social consensus more easily. What if we explicitly deployed the moral bases of educational decision-making rather than downgrading, understating or even concealing them? Assuming and deploying the value-ladenness of educational decision–making and renouncing the sirens’ song of the evidence-based educational decision-making imply changing the “vocabulary of motives” (Wright Mills, 1940) through which social acceptability is pursued. The legitimate repertoire of reasons to account for educational decisions, policies, and practices should shift from the “if X then y” logics and benefits/costs ratio, to a morally oriented vocabulary of motives.

There is a cost. Advancing the legitimacy of what I call moral accountability and axiological transparency implies expecting and accepting the unavoidable quota of conflict that any explicit reference to an axiological base of policies and practices may produce. Independently of any positioning toward their nature (essentialist realism vs. socio-cultural historicism), values are divisive particularly within a heterogeneous and pluralistic society. When “community” no more indicates a sociological category but rather a social principle ordering hyper segmented societies and producing increasingly intersectional identities, when individualized and children-and-family centered education are the pedagogical mantra, which community should we refer to when basing educational policies and practices on explicitly shared or to be shared values? Which community will lobby for which vocabulary of motives? Whose “good and right” will support, govern and account for collective educational decisions, policies and practices? These questions open a Pandora’s box, the one where avoiding surreptitious colonialism, cultural assimilation and normative socio-cultural models, risk producing a hyper-fragmentation of customer (or client?) tailored educational policies and practices.