Misinformation, Post-Truth Climates, Autonomy, and the Legitimacy of the Democratic State — II. How Misinformation Undermines Autonomy

Aubrey Beardsley — ‘The Black Cat’

This is the second in a series of four essays based on a period of writing towards the end of 2020. In this short series of essays I will introduce what I take to be differing types of misinformation, discuss the notions of individual autonomy and democratic engagement, and finally make comments on the legitimacy of modern democratic states and their relative obligations. I will ultimately argue that the legitimacy of modern democratic states hinges on the continued autonomous engagement of their populace, and that this autonomous engagement is diminished by misinformative expressions. One upshot of this is that if state-misinformation (i.e. instances where misinformative expressions have their basis in the actions or inactions of the state) does amount to significant losses in autonomy, then the state may actually serve to de-legitimise itself.

In this second instalment I aim to greater detail how the uptake of misinformation (and the beliefs, values, and actions which follow) can damage our capacity for autonomy; where ‘autonomy’ is understood as the authenticity of our ‘internal states’, and our ability to bring about outcomes which we desire.

i. A Definition of ‘Autonomy’

Apparently our autonomy is under threat all the time. Globally we hear that advertising and social media negatively impacts our autonomous development. We hear that those abroad suffer losses in both autonomy and liberty due to propaganda, religious dogma, or dictatorial regimes. Perhaps less severely, we hear people complain that their line of work doesnt afford them sufficient autonomy. Familiarity and continued exposure to terms like ‘autonomy’ can sometimes give us the illusion that we understand perfectly well what the term means, but it can be helpful at times to actually pause and reflect on what it is we mean when we talk of ‘autonomy’. It is only once we have some answers to these types of terminological or conceptual questions that we can look into how misinformation might undermine or inhibit the presence of this much employed term.

We can — for all intents and purposes — work with and develop a broad understanding of autonomy here. The purpose of this is not only for brevity (there are many substantial works devoted to the topic of autonomy; see Dworkin (1988), Christman & Anderson (2009), Conly (2014)), but by utilising a broad understanding of what it means to be autonomous we might better grasp the ways in which misinformation, in all its forms, can run counter to it. To consider someone as being autonomous we require certain conditions to obtain.

There is no universal agreement on these conditions, but — again, broadly speaking — we might plausibly assert that a first precondition for autonomy is what we can label ‘authenticity requirements’, and these relate to how ‘authentic’ our desires, beliefs, and valuations (i.e. our ‘internal states’) are, in the sense that they might rightly be labelled our own. One test for the authentic (and therefore autonomy-signalling) nature of our internal states is to consider whether, upon reflection, we are happy to possess and endorse such internal states and that the given internal states themselves cohere with our other valuations and beliefs about the world. To use an easily recognisable example, we might, for instance, enjoying smoking cigarettes, but find that smoking doesn’t cohere with our other beliefs and values about maintaining bodily health and setting good messages for our children. In instances like these we can propose that our ‘like’ of smoking is non-autonomous on some more fundamental level which accurately represents (to a greater extent) who we consider ourselves to be. We could even go so far as to say that when we smoke, we do so against our will or wishes.

Another set of preconditions for autonomous status we can understand as ‘competency requirements’. When we talk of somebody being competent we mean that they are often successful in some specific way or that they are capable of performing some task to a given standard (e.g. literary or sexual competency). In relation to autonomy, being ‘competent’ often equates to having the capacity for rational thought, self-understanding, and the freedom from oppressive factors in our lives which are likely to inhibit these. It is our ability to competently navigate ourselves through the world in a manner we deem fit and successful. These requirements then, align with our commonsensical intuitions about what autonomy is. Autonomy resides in having beliefs and values with are our own (authentic), and it is the ability to, allowing certain conditions, rationally (and competently) express these beliefs and values through our actions and behaviours in the world.

Now we have a working definition (even if it is a little rudimentary) of ‘autonomy’ which we may use to serve our current investigative purposes. Straightaway you may feel inclined to suggest that our consumption of misinformation has the ability to negatively impact and undermine both kinds of authenticity and competency conditions. I am of the same persuasion. Let us now turn to put some meat on the bones of the suggestion.

ii. Inauthentic Beliefs, Desires, Values, and Motivations

Seeing as we started with authenticity conditions, let us first ask how misinformation might damage autonomy in the form of these conditions. Take the hypothetical (though by no means unrealistic) example of an individual who we shall call Michael. Michael has been subject to a lengthy misinformational bombardment. The media he consumes, the political slogans he sees, and the regurgitations of these messages by people he knows have resulted in Michael forming several beliefs about his nation-state (the United Kingdom, say), the history of his nation-state, and the potential future of his nation-state. Exposure to and belief in misinformation has led Michael to conclude many things. One of Michael’s conclusions is that his ‘once great’ United Kingdom has slowly fallen from grace. Helping him internalise this belief (and others like it) is the fact that his experiental life is defined by financial difficulties, poor employment prospects, long wait times for medical appointments, and the physical dereliction of his local area. All things which he does not associate with his (flawed) past conception of the United Kingdom.

The misinformation Michael has been exposed to puts forward the idea that his misfortunes can be directly attributed to uncontrolled immigration policies. Michael is of the view that a steady but significant increase in immigrants entering the United Kingdom (some of whom do so illegally) has depleted job availability for himself and others like him. He thinks that the delays he suffers in his medical care are due to the resources of the National Health Service being siphoned by people who have emigrated to the United Kingdom for the exact purpose of free health care. Finally, misinformation has painted a picture of certain nationalities as being so barbaric and bestial that Michael believes their mere presence in his local area is a primary reason for its slip into disrepair. What we have here can perhaps best be described as a series of false beliefs and values which now assemble into a more overarching and fantastical narrative, a narrative which in turn influences Michael’s behaviour in numerous ways.

Michael’s non-autonomy resides in the fact that without the misinformation (in all its forms) which he has been exposed to, he wouldn’t have these internal states (e.g. his beliefs, views, and desires relating to the United Kingdom and the phenomena of immigration) to begin with. The reason that Michael’s valuations aren’t authentically ‘his’ is because of the process whereby they have been gradually and thoroughly implanted into him. Accordingly, even if Michael does offer some reflective endorsement of his beliefs (in the sense that he wouldn’t change them, or thinks they cohere with his fundamental self), what he is endorsing still remains inauthentic.

Admittedly we must be careful here not to whitewash all holders of similar beliefs as being similarly non-autonomous (this ‘patronising’ approach is, I believe, partly to blame for the massive divisions seen in recent political debates). No doubt there are many other instances where people are exposed to information (some of it false), and perhaps they are exposed to such (mis)information even in the indoctrinated style which Michael was. Some of these instances however will feature individuals which ultimately do nevertheless come to believe, agree with, and endorse on a properly authentic level the kinds of (mis)information they have encountered. In this sense, it is entirely possible that people with similar beliefs to Michael are autonomous. But there are also, I believe, many instances where we can rightly assume that an agent — such as Michael — would have reached different conclusions and value judgements were they not repeatedly exposed to misinformation which is intentionally tailored for uptake and acceptance. We can say then, without any trace of inconsistency, that in some (not all) cases like that of Michael, inauthentic (non-autonomous) internal states obtain.

iii. Authentic Beliefs and Motivations Gone Awry

Now let us turn to discuss our second kind of autonomous failing, namely, that related to competency conditions. To do this we should first imagine, contra the previous example, that Michael is to some extent autonomous with regards to authenticity conditions. Now of course nobody is formed in a vacuum, but Michael has developed his belief set in as individual a manner as is possible in a world with multiple exogenous influences on his thought and talk. He has, we might say, come to possess the relevant set of internal states (considerably) through his own volition. He endorses such states, and they authentically reflect him on some fundamental level.

Yet let us further stipulate that despite autonomy in this particular regard obtaining to some degree, when it comes to the topic of the United Kingdom exiting the European Union Michael remains particularly misinformed about the likely consequences of leaving, believing that boundless economic prosperity and swathes of job opportunities will follow. His comprehension is distorted, so that even when his valuations and motivations to vote ‘leave’ are autonomously formed and properly endorsed (i.e. authentic), when he comes to act on these internal states via his political participation something goes awry. The coveted (and predicted) effects do not come good. As most of us are now aware (if we weren’t before), with regards to Brexit this kind of scenario is all too true. We have seen economic decline, the weakening of the Pound, job uncertainties and mass redundancies, and ultimately the exact opposite of ‘taking back control’.

So how is Michael non-autonomous in relation to his competency conditions? Well, as O’Connor & Weatherall neatly put it: “If you believe false things about the world, and you make decisions on the basis of those beliefs, then those decisions are unlikely to yield the outcomes you expect and desire” (2019, p.6). As a result, due to the internalising of misinformation we see a sharp conflict between Michael’s actions and the outcomes he both intends and desires, and this failure to understand the most likely outcomes of his actions amounts to an absence of the kinds of competency requirements tied to the possible attainment of autonomous status.

Let us consider one further example where another hypothetical agent, Michelle, succeeds in her aims despite being misinformed. If Michelle shows signs of success — even in the face of misinformation — is she still incompetent? Say that Michelle looks to avoid becoming pregnant from the sexual intercourse she has. A series of bizarre posts on a forum which Michelle frequents has led her to believe that three antihistamines fulfil the same function as a more traditional (and medically verified) contraceptive ‘morning after’ pill. It isn’t impossible that Michelle manages to avoid pregnancy a fair few times and that she (wrongly) attributes this success to her antihistamine ‘contraceptive’, but — generally speaking — we would be confident in asserting that her conduct (of using antihistamines as contraceptives) is not conducive to continued success.

Why is this? As we saw in the example of Michael’s Brexit vote, when agents aren’t as lucky as Michelle it is often their false beliefs which determine the misalignment between their actions and how the world works. In this regard, the exposure to and uptake of misinformation (and the subsequent formation of false beliefs which follows) means that those exposed struggle to join the dots between their values (or internal states more generally) and the actions which express or realise these values in their lived experience. This lack in the competency (i.e. autonomy) department occurs even if the internal valuations under question (like Michael’s desire to improve the welfare of United Kingdom nationals, or Michelle’s motivation to avoid pregnancy) are ones which satisfy the criteria for being autonomously (authentically) formed.

The disconnect between belief and action caused by misinformation essentially means that we lack control, and instead — much like Michelle seeking to avoid pregnancy — remain unconsciously reliant on factors which are outside of our immediate jurisdiction (in other words, luck) working in our favour. Even when we, like Michelle, aren’t always circumstantially worse off when these types of misinformation-induced misalignment occur (though of course, we often are) between our internal states and our actions we still troublingly resemble the archetypal character of Greek tragedy, with our sovereignty or ‘destiny’ being independent of us, and in the lap of capricious and whimsical Gods. I think I speak with common sense behind me when I say that we, on the whole, don’t wish to be the characters in a real life tragedy.

But if misinformation is so damaging to our autonomy (in both senses discussed) then it seems correct to say that information is usually good for our autonomous capacities. Again, if misinformation contributes towards these specific instances of non-autonomy, then it seems that the way to offset these failures (and to move towards successes brought about by our own authentic self-governance) is through both information and knowledge, but exactly how much is needed? In light of this conclusion, we might worry as to whether the informational requirements for autonomy are actually unreasonably high.

Geopolitical events like Brexit and phenomena like global warming are incredibly complex affairs, and even experts in relevant fields like macroeconomics and climate science heavily rely on the informational input of others. If the picture is incomplete for people who devote their lives to understanding such complicated matters, should we really be surprised that there are many many informational gaps in our lives, and that these gaps contribute to losses in autonomy? In suggesting that information (and the absence or countering of misinformation) is required for autonomy, we might be left facing the uncomfortable conclusion that generally speaking most people aren’t autonomous about a great deal of things. If this further conclusion is accepted, we must ask whether certain informational requirements are too much of a demand for autonomous status, or perhaps whether misinformation really does play an abnormally potent autonomy-diminishing function.

I think the resolution to this issue lies in viewing our possession of autonomy as something measurable by degrees, a spectrum if you will. We can admit that we are more or less autonomous with regards to how certain beliefs, desires, and motivations interact with our comprehension of given phenomena, whilst at the same time we can maintain that there are significant thresholds for (and benchmarks of) autonomy. Running with this line of thought allows us to suggest that the more misinformation-uptake which occurs in our lives then the lower our chances of possessing authentic internal states, comprehensive abilities, and competency in seeing out our aims. From this, it readily follows that as our authenticity and competency conditions are weakened our degree of our individual autonomy is lessened also. This therefore means that a near omniscient epistemic or intellectual mastery of phenomena is not a necessary condition for autonomous values and actions, but the closer we move towards that idealised end of the spectrum, the more likely we are to be authentic and competent, and the better suited we are for autonomous status.

iv. Is Misinformation Always Detrimental to Our Autonomy?

We have now established two ways in which misinformation can lead to losses in autonomy. Firstly, when misinformation-uptake deprives our beliefs and valuations of their authenticity. Secondly, when misinformation effects our competencies by severing links between how we understand the world to be, our consequent actions, and the way the world actually is. This might be an appropriate time to question then whether misinformation is something which always runs counter to autonomy.

For instance, we might adopt the kind of thought suggested by Mill (2008) that the greater the amount of information disseminated the better circumstances presented for agents to test the mettle of their ideas in the quest for truth, and that true autonomous status requires precisely this skilful navigation of the informational morass. Nagel (1995, p.97) echoes a similar thought, arguing that opinions which dissent from established norms (a common feature of much modern misinformation is that it rallies against establishment views) should not be suppressed in favour of retaining orthodoxy, and that we, the audience of such expressions, can perhaps best achieve autonomy by deciding how to appropriately utilise (mis)information in the making up of our own minds.

Lastly, there is not always a guarantee that truth (or information which tracks the truth) is conducive to autonomy. It may well be the case that our possession of certain facts in conjunction with the kinds of beings we are is damaging for our epistemological conduct. History shows that making certain pieces of (genuinely true) information available to agents like jurors (e.g. that a defendant has acted in an unsavoury manner in an unrelated instance many years ago) can impact their decision making in unhealthy ways, and it is thoughts of this ilk which prompt Goldman’s advocation of ‘epistemic paternalism’ in certain legal scenarios (1991, p.118). Even if one takes issue with claims like those put forward here by Goldman, it cannot be entirely ruled out that given the types of finite, prejudiced, and otherwise epistemically imperfect beings that we are, unrestrained information is not always a prerequisite for autonomous beliefs, judgements, and actions.

Perhaps the best way to respond to these briefly sketched counterexamples questioning the links between information and autonomy (and by extension, the links between misinformation and non-autonomy) is to again refer to the notion that autonomy is something permitted in degrees and as a characteristic that requires some threshold being reached. Keeping this understanding of ‘autonomy by degrees’ in mind we can argue that even though misinformation-uptake doesn’t necessarily equate to an agent being non-autonomous (nor information-uptake necessarily equating to autonomy) these instances of misinformation-uptake generally work against the achievement of the aforementioned autonomous threshold.

Under this framework, it seems to follow that we can readily permit examples where an individual is exposed to misinformation and in possession of certain false beliefs, but nevertheless remains capable of achieving the status of autonomy in both general or more phenomena-specific senses. As touched upon earlier, we must also be conscious, whilst conscientiousness is of the hour, not to equate a certain belief set with an agent’s non-autonomy. Belief X might often be symptomatic of Y (where Y is a failure of rationality, mental soundness, or autonomy, say), but a converse scenario (where X is present but not Y) shouldn’t be considered unintelligible solely in virtue of what we usually find to be the case (Feinberg, 1971, p.117).

There are plenty of belief sets where we can reasonably predict that the agent in possession of said belief set is inhibited in their capacity for autonomy — take a Saudi Arabian woman’s belief that she is inferior solely due to her gender, or the values of an ardent subscriber to Alex Jones’ ‘Infowars’ — but their non-autonomy should not be something asserted a priori. With this in mind, we should be open to the fact that one’s autonomous status must be evaluated on a somewhat individual basis.

A final lingering thought is the following: if not all instances of misinformation-uptake (and any inauthentic beliefs or incompetencies which follow) automatically diminish autonomy, then is there any way of predicting which instances do? Broadly speaking I think two determinate factors must be considered here. Firstly, we must consider the content of the misinforming communication, the manner in which it has been disseminated, and ultimately how worldview-distorting its uptake and corollary beliefs frequently are to people exposed and commited to such misinformation.

As a bit of a truism, something may only be worldview-distorting if it is in relation to some given worldview, so from our first consideration it follows that the second factor at play is more contextual, namely, how these variables interact with a particular individual and their worldview. To employ a now familiar example let us take the case discussed in the previous essay concerning how algorithms might be employed to generate misleading survey results about policy approval ratings. Even if one undergoes belief-formation in light of misleading information surrounding the approval of a given policy, it might not largely distort our picture of the world (though of course it will probably play a distorting role to some extent). For another agent however, someone who has been cognitively primed for these types of misleading political messages — or who has a greater epistemic narrative or belief corpus that the expression fits into snuggly — might find that their worldview — though no doubt already skewed — has been warped to a far greater extent by what is the same expression of misinformation.

The threat of misinformation on autonomy therefore depends firstly on content (the type of misinformation being relayed and the manner in which it is relayed) and secondly on context (precisely who is hearing it). No doubt trends and predictions can be made with regards to how particular pieces of (mis)information are likely to impact a certain demographic, but we still come back to the thought that precise links between misinformation and one’s autonomy are something that must — at least to some degree — be understood on a more personal basis. All of this however, does not detract away from the previously argued link between misinformation-uptake and general losses in autonomy. We may still readily assert that when an agent exemplifies certain kinds of uptake and belief formation, then their internal states are likely to endure some depreciation in authenticity. Furthermore, we can claim that in similar circumstances of misinformation-uptake the agent will be less competent (even if their internal states are autonomously formed) in seeing out their aims through their interactions in the world.

Given that we have, I hope, established a working definition of ‘autonomy’ and detailed how misinformation serves to undermine it, we are now in a position to isolate precisely what exactly is problematic about these losses in autonomy when the source of misinformation is the state. This is the primary aim of the section that follows.

References:

• Christman, J. & Anderson, J., ‘Autonomy and the Challenges to Liberalism: New Essays, Christman, J. & Anderson, J. (eds), Cambridge: Cambridge University Press, (2009).
• Conly, Sarah., Against Autonomy: Justifying Coercive Paternalism, Cambridge, United Kingdom: Cambridge University Press, (2014).
• Dworkin, G., The Theory and Practice of Autonomy, Cambridge: Cambridge University Press, (1988).
• Feinberg, J., ‘Legal Paternalism’, Canadian Journal of Philosophy, Vol.1, №1, (1971), pp. 105–24.
• Goldman, A. I., ‘Epistemic Paternalism: Communication Control in Law and Society’, The Journal of Philosophy, Vol.88, №3, (1991), pp.113–31.
• Mill, J. S., On Liberty and Other Essays, United Kingdom: Oxford World’s Classics, (2008).
• Nagel, T., ‘Personal Rights and Public Space’, Philosophy & Public Affairs, Vol.24, №2, (1995), pp.83–107.
• O’Connor, C. & Weatherall, J. O., The Misinformation Age: How False Beliefs Spread, Yale: Yale University Press, (2019).