Introduction

Deception is an omnipresent and pervasive feature of human interactions—in everyday lives, we all tell numerous white (and not so white) lies in order to show our best side to the world. Public personae such as actors and singers present themselves using highly mediated images and videos in order to look young, beautiful and desirable; politicians send carefully crafted messages that show them in (what they believe is) the best possible light. Looking at individuals, the psychologists Whitty and Joinson argue ‘lying is a part of daily life. The Internet has simply provided a new place for individuals to lie’ (2008, p. 56). However, the Internet has not merely enhanced individual human tendency for deception. Defined as ‘circumstances in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief’ (Oxford Dictionaries 2016), post-truth is a far-reaching social concept which threatens the traditional concepts of knowledge, decision-making and politics.

According to Rider (2017), ‘in order to understand the idea of “post-truth” or “post-fact” politics, it is tempting to focus on the lie. I think there is reason to resist this temptation. There are a plenty of utterances that fall in between’. This chapter examines such utterances and defines post-truth as a curious phenomenon that resides in between truth and lie, emotion and reason. Based in a philosophy of trust and analyses of digital trust, it defines post-truth as a poisonous public pedagogy and seeks response in a new, whole-rounded critical pedagogy of trust.

The Curious Divorce of Truth and Trust

Hancock (2007) defines digital deception as ‘the intentional control of information in a technologically mediated message to create a false belief in the receiver of the message’. This definition consists of three important parts: (1) deception needs to be deliberate, (2) deception needs to be addressed at someone, and (3) deception needs to be technically mediated. Hancock furthermore classifies digital deception into identity-based deception, which consists of faking identities, and message-based deception, which consists of faking the content of communication. Using these helpful classifications, he studies the psychology behind various forms of online deception from fake e-mails to fake online dating profiles.

Let us apply Hancock’s definition to a recent example of post-truth. On 18 February 2017, ‘Donald Trump appeared to invent a terrorist attack in Sweden during a campaign-style rally in Florida on Saturday, inviting questions that he may have confused the Scandinavian country with a city in Pakistan’ (Topping 2017). Trump’s deception is clearly message-based, because it consists of providing false content of communication. However, it is impossible to determine whether the deception was deliberate and who is the exact addressee of the deception. The act of deception took place in a public talk, yet the majority of its recipients received information through one or another technically mediated medium such as television and the Internet. Now let us take a brief look at the consequences of Trump’s talk. On the one hand, the video recording clearly shows that Trump’s deception has produced the desired emotional response from the audience. On the other hand, many people read news reports revealing the deception, and some of them may have formed different opinions based on that revelation (see, for instance, readers’ comments under the article which reveals the deception) (Topping 2017).

The majority of people received Trump’s message online, so a possible route for analysis is the psychology of online deception. Drawing from Hancock (2007), Whitty and Joinson describe the truth-lies paradox:

If the technology itself is both an enabler of honesty and deceit, one needs to look beyond explanations rooted in technological determinism in order to fully understand people’s behaviour online. Specifically, we would argue that to understand the nature of truth and lies online, one needs to look at the context in which people act alongside the person themselves. We would also argue strongly that truth and lies are not mutually exclusive, and that in much online interaction people are strategically managing their online identity to meet both their own goals and the expectations of the other. While doing this, they are also balancing their actions with the norms for the site or community in which they are active (Whitty and Joinson 2008, p. 143).

Finally, Whitty and Joinson place ‘trust as central to the truth-lies paradox’ (Whitty and Joinson, p. 143).

For Hancock (2007) and Whitty and Joinson (2008), truth is dialectically intertwined with trust: I trust you, because I expect that you are telling the truth. In the context of post-truth, however, it is almost completely irrelevant whether the alleged terrorist attack in Sweden really happened. Appealing to emotion and the personal beliefs of his audience, Trump reinforced their widespread fear of Islamic terrorism as a rising threat to the Western world. Then, Trump used the emotional resonance with his audience in order to create trust that he is the right person for resolving that problem—in this context, ‘minor’ issues such as truthfulness of his claims are secondary.

What caused this curious divorce between truth and trust? In her insightful explanation, Sharon Rider takes as her point of departure Peter Thiel’s remarks at a meeting of the National Press Club in Washington DC: ‘I think one thing that should be made distinguished here is, the media always is taking Trump literally. It never takes him seriously, but it always takes him literally. I think a lot of the voters who vote for Trump take Trump seriously, but not literally’ (cited in Rider 2017). Rider’s conclusion is not simply that voters are willing to ‘forgive’ a lie or two to their favourite candidate. Rather, she argues that post-truth creates an intellectual universe wherein the message conveyed is that evaluation according to criteria of truthfulness is irrelevant.

If Thiel’s analysis is correct, Trump’s victory is not due primarily to the unleashing of hitherto suppressed sexist, racist, homophobic, anti-Semitic sentiment among white male voters, but in an intuition that Trump’s bombastic and seemingly incoherent statements are not statements at all. (…) Lying assumes that what is being conveyed is content, but on Thiel’s interpretation, Trump is not conveying content. He’s not talking about policy, but rather about an attitude toward policy (Rider 2017).

Obviously, this fundamental rejection of truth results in stark consequences. ‘When we retreat from the responsibility of establishing and recognizing a minimal common ground for debate and decision, we implicitly dismiss the possibility of democratic forms and processes’ (Rider 2017). By divorcing truth from trust, post-truth does not bring about yet another form of the lying politician modelled in the image of Donald Trump. Instead, it fundamentally threatens traditional conceptions of knowledge, decision-making and politics.

Post-truth and the Philosophy of Trust

The post-truth divorce of truth and trust, combined with the rejection of truth, brings into foreground the questions of trust. In the following discussion, therefore, I will examine the relationships between truth and trust using four main themes: (i) the nature of trust and trustworthiness, (ii) the epistemology of trust, (iii) the value of trust and (iv) trust and the will (McLeod 2015).

  1. (i)

    Trustworthiness is a human property, while trust is an attitude towards other people. Based on that distinction, Macleod identifies four main requirements for establishing trust.

Trusting requires that we can (1) be vulnerable to others (vulnerable to betrayal in particular); (2) think well of others, at least in certain domains; and (3) be optimistic that they are, or at least will be, competent in certain respects. Each of these conditions for trust is relatively uncontroversial. There is a further condition which is controversial, however: that the trustor is optimistic that the trustee will have a certain kind of motive for acting (McLeod 2015).

In the age of the digital reason, we are more vulnerable to betrayal than ever—information can be easily manipulated; identities and messages can be easily falsified. At the same time, however, we are less vulnerable to other forms of attacks; for the most part, technology-mediated hate messages are less hurtful than physical violence.

People strategically, and often unconsciously, manage their online identities in order to meet social expectations (Whitty and Joinson 2008, p. 143), so thinking well of others and believing in their competence can be based on dubious information. Furthermore, even the most correct information is useless without correct interpretation. For instance, my colleague from information services lists the following abbreviations in his e-mail signature: MCP, MCTS, MCSA, MCITP and MCT. The abbreviations look impressive, so I took the effort and found out that they represent a list of Microsoft’s certificates (Microsoft 2017). This indicates that I should trust my colleague’s competence—after all, he is certified by one of the largest information technology companies in the world. Being a technological novice, however, I do not really understand the meaning of these certificates, so I still do not know whether they will help my colleague to resolve the problem with my computer. Macleod’s fourth requirement for trusting, the trustee’s motive for acting, is also problematic. In my work environment, I can trust that my colleague is motivated to resolve my computer problem because he gets paid for the job. However, if I seek help from an online company, I have no idea who sits behind the other screen and why. Trust and trustworthiness have never been easy, yet it is safe to say that the digital age has brought about an additional level of complexity—the age of widespread information has brought about increasing difficulties in trusting that information.

  1. (ii)

    The central question of traditional epistemology of trust is: ‘Ought I to trust or not?’ (McLeod 2015). Since the dawn of humankind, the philosophical branch of scepticism has brought about valuable insights into this question. In scepticism, however, trust is dialectically intertwined with truth—and post-truth explicitly rejects this relationship. In a post-truth environment, therefore, pragmatist and rationalist approaches focused on problem-solving and action provide a much better fit. For instance, I can meet my brother and unconditionally believe what he told me in person—these things happen between siblings. However, Faulkner shows that trusting and being trustworthy cannot be explained merely by reference to ‘person’s beliefs and desires’ or teleological considerations, and claims that ‘in trusting one takes on commitments, not merely to act in certain way, but also to premise one’s practical reasoning on a trust-based view of the interaction situation’ (Faulkner 2014, p. 1795). Depending on the context, my trust-based view can change. I unconditionally believe my brother, yet I can never be completely sure about truthfulness of his Facebook posts and e-mails because his accounts might have been hacked. It is easier (and more common) to hack social media accounts than e-mails, so my trust in brother’s Facebook post is lesser than my trust in brother’s e-mail. As this example illustrates, rationalist context-specific approaches to traditional epistemology of trust are suited to the age of the digital reason.

Traditional epistemology of trust is primarily concerned with the individual. However, post-truth is an inherently social phenomenon, so it should be examined using the approach of social epistemology. For instance, advocates and deniers of climate change often use the same empirical datasets in order to arrive at opposing ‘scientific truths’ (Gleick 2007). Certainly, the question whether human beings cause global warming has only one true answer—so one of these groups uses pseudoscience to prove own ideological positions. However, what happens when both sides of the debate possess certified ‘experts’ with proper degrees and credentials? And what happens if we cannot easily identify political, economic and other interests behind the competing ‘truths’? According to Goldman and Blanchard (2016),

A fundamental problem facing the layperson is that genuine expertise often arises from knowledge of esoteric matters, matters of which most people are ignorant. Thus, even when a layperson listens carefully to someone professing great expertise, the layperson may be at a loss to decide whether the self-professed expert merits much trust.

For laypersons faced with stark disagreement between the experts, Goldman and Blanchard list several strategies for finding the truth: (1) ‘to arrange a “debate” between the self-professed experts’; (2) ‘to inquire which position endorsed by one of them is most common among all (professed) experts’; and (3) to compare ‘their respective track-records: how often has each expert correctly answered past questions in the domain?’ (ibid). However, these methods are sometimes indecisive and sometimes hard to convey, so laypersons naturally incline to scientific ‘truths’ which provide a better fit to their overall emotion, personal belief and underpinning motifs and interests.

The testimony of experts is just one possible spin-off of a more general problem of testimony. Social epistemology inquires into the epistemic nature of collective agents, issues pertaining to scientific knowledge, democracy, free speech and ethics. While traditional epistemology illuminates post-truth within a rationalist framework, social epistemology provides various useful tools for analysing the collective production of knowledge and its social consequences.

  1. (iii)

    Trust is the bread and butter of human civilisation. According to Macleod, ‘without trusting or being trusted in justified ways, we could not have morality or society and could not be morally mature, autonomous, knowledgeable, or invested with opportunities for collaborating with others’ (McLeod 2015). In the age of the digital reason, many traditionally individual activities are increasingly transforming into the realm of collective thinking and the social production of knowledge (the transition from traditional humanities to digital humanities is a typical case in point) (Wark and Jandrić 2016; Peters and Jandrić 2017; Jandrić 2017a, b). Thus, trust becomes increasingly important for almost all kinds of knowledge development.

  2. (iv)

    Trust is cultivated through a history of interactions between the trustees—one cannot wake up one morning and decide to trust someone. Therefore, philosophers have defined trust in terms of belief, emotion and various combinations thereof (McLeod 2015). A decisive take on the relationships between trust and the will is far beyond the scope of this paper, yet the non-voluntarily nature of trust as belief and the complexity of trust as emotion are essential for understanding the nature of post-truth. Post-truth does not care about truth, because it is emotional. Post-truth is non-predictable, because it is not rational. Post-truth strongly influences people, because it appeals to basic human instincts. As can easily be seen from Donald Trump’s presidential campaign, the emotional, the irrational and the instinctive cannot be counterbalanced with truth and reason. This conclusion does not imply that truth and reason are unimportant—in the face of blatant lies and stupidity, most people will respond negatively. But this negative response is often instinctive and emotional, which is to say, irrational. Thus, the response confirms the content of the post-truth message, namely, that emotion and instinct are often more powerful than truth and reason.

Post-truth and Digital Trust

Digital information and knowledge heavily depend on trust—they require us to trust the machine, the people behind the machine and the logic within the machine. Digital machines require input information, which can be true or false. With adequate input information, their inner workings consist of numerous adding operations between two distinct electronic states described as zero and one. However, human beings are physically unable to comprehend long lines of numbers, so computers execute many levels of translation between the source code and programming languages. Some computer programs, called algorithms, have certain abilities for personalised, autonomous-like action. For instance, recommender engines such as Amazon.com use buyer’s history of browsing and purchases in order to advertise products that might be of interest. This automatic translation and algorithmic manipulation is generally straightforward, and does not raise significant trust issues. For as long as computer programmers do not purposefully implement malicious code, such systems are generally safe to use.

As of recently, however, algorithms are becoming increasingly complex; moreover, the computer industry is strongly oriented towards the integration of various algorithms. For instance, browsing the Internet using Google collects data about our interests; data from Google Mail reveal our communications; buying on Amazon.com reveals our shopping habits; participating on Facebook reveals our social networks; using Google Maps reveals the history of our physical movement; and using porn websites reveals our sexual fetishes. When cross-referenced, this data ceases to be innocent, and this brings about the emerging field of algorithm studies. The integration of algorithms poses significant challenges, especially regarding the ownership of data and the manipulation of data.

Scientists of the past believed their books and logarithmic tables; scientists of today believe their journal databases and calculators. Arguably, there is no significant difference between publishing an ill-informed book on paper and an ill-informed article online. However, this is where similarities end. In the case of physical books, one could be reasonably sure that the presented text is original; in the case of online articles, one can never be sure whether the content was manipulated by a third party on the long road between publishing and reading. Furthermore, non-manipulated articles can contain true or false statements, authentic or misleading statistics, diagrams or images. Given that most people today find their information online, the trustworthiness of our online searches deserves more focused attention.

Analysing Trump’s talk in Florida (Topping 2017), it is easy to check whether a terrorist attack indeed happened in February 2017 in Sweden—such an event would surely provoke wide media coverage. Using standard online search tools, however, some truths are much more difficult to verify. For instance, Adolf Hitler’s Propaganda Minister in Nazi Germany Joseph Goebbels allegedly said: ‘If you repeat a lie often enough, people will believe it, and you will even come to believe it yourself’. A simple Google search on this sentence returns an astonishing number of 9,570,000 pages; as an exact quote, it returns 937 results.Footnote 1 Reviewing more than 9 million websites is practically impossible for a single person, so the best thing one can do is to look into the first few web pages and hope that Google has picked the most relevant ones. However, we probably do not need all the results anyway: even a brief look at the most popular web pages indicates a serious lack of evidence that Goebbels really made that statement. To the contrary, it is attributed Goebbels, Hitler and others in the randomly revised websites.

The majority of researched pages quote the sentence without questioning its factual correctness. However, some of the found pages do question its validity. For instance, author of ‘Think Classical’ (2016) blog claims that the quote ‘is apocryphal, and there is no evidence that he ever articulates such a principal. It is high irony indeed that the apocryphal quote has been falsely attributed to so Goebbels so many times that people have come to believe it’. Symptomatically, the blog does not reveal the full identity of its author and does not list (all) the used sources—therefore, this assertion is also far from a proven fact.

Another famous adage with unclear source(s), Segal’s law, says: ‘A man with a watch knows what time it is. A man with two watches is never sure’.Footnote 2 On the Internet, Segal’s law multiplies to stunning proportions. Answering my simple query, Google has offered 9,570,000 answers—but none of the web pages I was able to review have provided actual ‘facts’. So how do I determine whether Goebbels really made that statement? Perhaps, using more sophisticated search methods, I would eventually find an answer in a reliable source such as an academic book or article. However, if the topic has not been covered in reliable sources, my only recourse is to do my own research on primary data, which could of course take years.

There is no such thing as digital trust—only a mash-up of visible and not-so-visible data and algorithmic functions. Therefore, the trustworthiness of digital sources is always established in relation to the nondigital. Online credit card frauds are resolved by tracing physical goods, online identity theft is resolved by tracing physical documents and trust in online documents (such as journal articles) is established by the reputation of their publishers. The complexity and murkiness of digital technologies widen the ambiguous space between truth and lie, thus providing fertile ground for post-truth. However, the problem of post-truth is not exclusively digital, and understanding it takes us beyond the digital realm.

Post-truth as a Public Pedagogy

In information sciences, the content of the human mind is commonly classified according to the so-called data–information–knowledge–wisdom (DIKW) hierarchy (Fig. 1). ‘Data are symbols that represent the properties of objects and events. Information consists of processed data, the processing directed at increasing its usefulness’ (Ackoff 1999, p. 170). ‘Knowledge builds on information that is extracted from data’ (Boddy et al. 2005, p. 9); depending on philosophical traditions, it can take numerous shapes and definitions. Finally, ‘wisdom is the ability to act critically or practically in any given situation. It is based on ethical judgement related to an individual’s belief system’ (Jashapara 2005, p. 17–18).Footnote 3

Fig. 1
figure 1figure 1

The data–information–knowledge–wisdom (DIKW) hierarchy

Full size image

According to Rowley, ‘There is a consensus that data, information and knowledge are to be defined in terms of one another, although data and information can both act as inputs to knowledge. This consensus reaffirms the concept of a hierarchy that links the concepts of data, information and knowledge’ (Rowley 2007, p. 174). As can easily be seen from various analyses in this chapter, post-truth ignores truth at all levels: data is falsified, information is misprocessed and knowledge is distorted. Post-truth data, post-truth information and post-truth knowledge inevitably lead to post-truth wisdom. This conclusion has an important temporal dimension. Data, information and knowledge describe what was and what is—they are focused on past and present. However, wisdom provides guidelines for human behaviour—it looks straight into the future. Therefore, post-truth is a poisonous public pedagogy oriented towards raising future generations of people with distorted worldviews, opinions and ethical judgements.

The current critical media response to post-truth predominantly seems to consist of revealing lies and fallacies (Rider 2017). However, this study indicates that post-truth fundamentally rejects the criterion of truth, and thrives in the curious space between truth and lies; based on trust, it easily absorbs factual discrepancies and even blatant lies. The complex relationships between truth and trust can be analysed using Choo’s analysis of transformations between signals, data, information and knowledge. According to Choo (2006: 132), physical structuring of signals and data precedes cognitive structuring of information, which in turn precedes structuring of belief. Each of these stages provides more opportunity for human agency, which closes the loop between the structures. Some scenarios for this loop are as follows:

  1. 1.

    False data will produce false information will produce false knowledge—in turn, false knowledge will produce false interpretation of new (correct and false) information and data.

  2. 2.

    Correct data will produce false information will produce false knowledge—in turn, false knowledge will produce false interpretation of new (correct and false) information and data.

  3. 3.

    Correct data will produce correct information will produce false knowledge—in turn, false knowledge will produce false interpretation of new (correct and false) information and data.

  4. 4.

    Correct data will produce correct information will produce correct knowledge—in turn, correct knowledge will produce correct interpretation of new (correct and false) information and data.

The loop needs to be expanded by two improbable, but possible scenarios: false knowledge can produce correct interpretation of information and data, and correct knowledge can produce false interpretation of information and data.

The loops listed are based on linear relationships between data, information and knowledge; in reality, however, relationships between these elements are bidirectional and networked. Furthermore, the analysis could be refined by adding questions pertaining to signals and agency. However, even this simplified analysis of relationships between truth and trust clearly explains the futility of typical responses to post-truth focused on revealing lies and fallacies. In order to counterbalance the poisonous post-truth pedagogy, we need a critical pedagogy of trust that pays equal attention to data, information and knowledge.

Conclusion

Post-truth is a complex mashup of signals, data, information, knowledge and wisdom; truth and deceit; fact and emotion; reason and instinct. These concepts and forces have always marked human existence, yet the digital age made their mutual relationships increasingly complex. Truth and reason are as important as ever, yet post-truth thrives in an ambiguous space between truth and lie, reason and instinct—in this space, truth gets replaced by trust. Mainstream analytic philosophy seeks individual solutions in rationalist approaches to trust, and social epistemology seeks collective solutions in areas such as the problem of testimony, social epistemology of science, ethics and democracy.

The era of post-truth came alongside the era of digital reason, yet the trustworthiness of digital sources can be established only in relation to the nondigital. Trust is the main prerequisite for digitally enabled collective intelligences, yet the inherent untrustworthiness of digital technologies indicates that we should place more value on trust in other human beings. Trust is cultivated from emotion and belief, yet it results in decisions about objective truth. Trust links our past and present (represented by data, information and knowledge) and our future (represented by wisdom). In this way, post-truth becomes a poisonous public pedagogy that can be counterbalanced only by a fully developed critical pedagogy of trust.