In the aftermath of the Global Financial Crisis (GFC), while dealing with the dramatic consequences, electors and their representatives started to search who was responsible. The GFC was a complex phenomenon in which many actors, communities, corporations and governments were involved. Yet, some appeared to have been more active than others and roles were not the same. Rating agencies had given many products there full approval (the famous “AAA”) before realizing how problematic they were, i.e. CDO carrying subprime mortgages. When confronted in Court and in Congress about their “reckless disregard for the truth” (The Economist, Sept. 6^th 2007), credit rating agencies used the First Amendment of the US Constitution that protects freedom of speech:

• “The ratings … are and must be construed solely as, statements of opinion and not statements of fact or recommendations to purchase, sell, or hold any securities“ (The Financial Crisis Inquiry Commission, Jan. 2011).

Since then, this argument has not always been enough and credit rating agencies have been held responsible to different extents depending on the contexts. We should not use them as an easy scapegoats despite their prominent role.

What I will focus on here is the impact these specific “opinions” had on the markets. If ratings are merely opinions of institutions that revealed themselves to be sometimes (widely) inaccurate, how come they still have such an influence when rating sovereign debts? Beyond the use of these ratings by business leaders and politicians, the interesting phenomenon here is the performativity of the speeches at play in this communication. Moreover, the fact that they are said to be “opinions” adds a problematic layer.

• What is an opinion?

When experts starts defending themselves by saying their expertise is, end the end, merely an opinion, we should question their very expertise since opinions is not knowledge or science, nor is it as strong as beliefs and faith. Opinions are inherently subjective, whereas experts tend to present themselves as objective, or detaining an objective expertise…

The etymology of “opinion” does not rely on an independent critical thinking since it comes from “opinari” which means to opine, to suppose, to judge. According to Plato (The Republic, VI), opinions are made of beliefs and illusions; they constitute what is projected on the wall of his Cavern, maintaining us within the confine of our prison. Opinions are thus opposed to knowledge and science. One can find this idea developed further in Kant’s Critic of Pure Reason in which opinions are considered as a specific type of belief. Definition of belief and opinion. Characteritics of opinion. Issues with opinion, persuasion and communication.

Opinions can be construed as an obstacle to knowledge. According to Bachelard, “opinion is by its nature always wrong; opinion thinks poorly – it does not think, it translates needs into knowledge. Nothing can be founded on opinion; it must be destroyed. It is the first obstacle to be overcome” (La formation de l’esprit scientifique). Translating needs into knowledge is interesting (and almost ironic) considering the fact that rating agencies are paid by the institutions, which produce and sell the rated financial products…¹

Although, since opinions appear to be so fragile, how come their production and diffusion came to have such a dramatic impact?

• The performativity of opinions

When kept to ourselves, opinions don’t matter and have very little impact. And we should do as such according to Kant… It is when communicated that opinions become performative: they produce something, they have effects and consequences. Especially if the source of this opinions is a well-known actor or a trend-setter.

To analyse the impact of ratings produced by Moody’s, S&P and Fitch, one could use Austin’s three-levels model. According to Austin (1962), speech acts can be understood through several layers of meaning:

• what is actually said (locution)

• what the speaker tries to achieve through the speech (illocutionary force)

• What impact the speech actually has on the audience (perlocutionary effect)

Consequently, beyond the credit agencies’ discourses synthetized in their ratings (from “AAA” to “junk”) and the solidity of their models, beyond the fact that they are now presented as “opinions”, their simple communication have tremendous effects. By the end, locutions should not be our point of focus. We should try to unveil their illocutionary forces: what is at stake behind this rating locution? And we should start analysing their perlocutionary effect: how do the communication of these ratings impact us?

Ratings are speeches, discourses in the forms of letters and numbers that pretend to capture the infinite complexity of the world. According to me, the important question is not whereas they are reliable or not. The fascinating phenomenon is how something that is now defined as opinions by their very producers comes to have such an impact on political and business leaders but also on the “public opinion”?

Yet, one should ask, what is this public opinion?

• On the issue of public opinion

In 1973, Pierre Bourdieu wrote a very interesting piece in the review Les Temps Modernes called: Public opinion does not exist. In this, he questions three fundamental assumptions of opinion polls: everyone can have an opinion on any given subject, all opinions are equivalent, and a consensus exists on which questions should be asked at a given moment. In doing so, he does not question the technical nature of these polls (representativeness, statistical models, etc.) but rather what one could call their epistemology.

A first bias in these poll is their very construction since they usually produced in response for an order by a person, a corporation or an institution. Consequently, questions and possible answers will be framed explicitly or tacitly; one should note that the verb “frame” is used, not “distorted” or “manipulated” which would imply voluntary bias. To Bourdieu, the existence of opinion polls is usually subordinated to political requests, and therefore is political in nature. Consequently, “public opinion”, as usually synthetized by a percentage in newspapers, is an artefact produced at a given moment that does not represent the dynamic forces on which it relies. Considering how polls and statistics are used in political speeches, they are, according to Bourdieu, political artefacts used to legitimize, or delegitimise, plans or decisions.

Moreover, these artefacts create a consensus effect that ignores non-answers (often erased in statistical models) and aggregates subtle positions in broad categories (satisfied vs. unsatisfied, pros vs. against, supporters vs. enemies). Or Bourdieu, the existence, structure, analysis and presentation of this so-called “public opinion” should be scrutinized starting with who ordered the poll and the intentions behind it.

According to Bourdieu, answering such polls and questions requires a political competency that one does not necessarily have on every issue all the time. We are not all competent on every subject and therefore merely have weak opinions (pleonasm as we now know) that barely deserve being expressed, not to mention statistically analysed.

Opinions, polls will produce a “public opinion” that is not “right or wrong”, nor “true or false”. It is a fiction that only has an effect through the performativity of communication. “Public opinion” is a political artefact that can be used in discourses and other speech acts in order to convince, legitimize, attack or defend. They impose what Bourdieu calls a “dominant issue” chosen by their producers and diffusers. Agreeing with it or being supported by it does not mean anything. Yet, just as financial ratings, these opinions can have dramatic perlocutionary effects if we believe in them…

• On the dark side of opinions’ performativity

Opinions, as the weakest form of belief according to Kant, are widely recognized as subjective forms of judgement that barely deserve publicly mentioning. Going further, I would state that the performativity of discourse renders the aloud expression of opinions is extremely problematic. This is in no way an opposition to the freedom of speech, but rather a reminder if the importance of its use not to express any weak judgement on any subject all the time.

Expressing one’s opinion (as opposed to beliefs, analysis, faith or knowledge) is performative in the sense that it allows for our inner private world – res cognitans – to have an effect on the world surrounding us – res extensa (Kjellberg & Helgesson, ). In doing so, we impose the weakest form of our judgements upon the world, not only as logorrhoea but also its effects. As Callon (1998) showed, the discourse of economic theory does not only describe markets, it shapes them. The same performative dynamics apply to our opinions on a varying scale depending on the extent of the diffusion of our opinions. Pollsters and credit rating agencies have shown how far and how deep these opinions can go. Consequently, since discursive statements have a determinant effect on how practices and activities can be implemented (Hardy, Palmer & Phillips, 2000), opinions should be expressed, heard and analysed with extreme caution. Even better, they should be reflexively examined to become stronger, richer judgements and statements before even being valued, used and imposed to others.