California State Senate will soon vote on a Bill prohibiting dissemination of unverified claims of fact on social media. A similar law is already in force in Malaysia (NYT) with France and Belgium expected to follow suit (CNBC). As a consequence, fact-checking will become the condition of participation in the public discourse. In this article I formulate methodology for fact-checking of any statement or source, including the fact-checkers themselves, without having access to all the relevant information.
Senator Richard Pan is the author of Senate Bill 1424, informally known as the Online False Information Act, which, if passed, would require any person who operates a social media Website with a physical presence in the State of California “to develop a strategic plan to verify news stories shared on its Web site” including “utilization of fact-checkers (…) and placing a warning on a news story containing false information.” The Bill does not specify who the fact-checkers are, but the new law would probably involve creation of an administrative body tasked with accreditation and licensing of fact-checking organisations, or mandatory adherence (without evidence to the contrary) to the information published by the accredited news networks. It is not surprising that the networks are dead-silent about this proposal. After their ‘fake-news’ meme has spectacularly backfired (WaPo), revealing pervasive public distrust of the networks and ultimately resulting in their widespread ridicule, the accredited news networks stand to benefit immensely from the proposed legislation by becoming de facto fact-checkers, with the capacity to censor, or at least moderate, any alternative viewpoints.
Mandatory reliance on approved fact-checkers in the public discourse is alarming for a number of reasons, affecting both the freedom of speech and the freedom of thought, but I will not dwell on the negative aspects here. Instead, I focus on how to fact-check any purportedly factual information or source, including the fact-checkers themselves.
Facts are just those things that truly exist or events that have actually happened (ontological description), or true pieces of information about things or events (epistemological description). One important feature of factuality of any kind is coherence – facts are causally connected and systemically consistent; if something is not causally connected and consistent with all other facts, then it is not a fact. The sense of Reality (or Truth) precludes contradictory facts: nothing can both be (true) and not-be (true) at the same time and in the same respect (The Law of Non-Contradiction). From this fundamental principle we can derive the first rule of fact-checking: no claim of fact shall be taken as true in its own right.
A consistent account of the ontological/epistemological distinction is crucial to effective fact-checking. While knowledge aims at grasping reality as something independent of any subjective point of view, what we take for practical facts (events or phenomena) are just beliefs based on essentially limited points of view, but formulated in adherence to conventions, categories and principles we take to represent the structural conditions of Truth. All true claims of practical factuality are then just ‘rational’ beliefs supported by first-hand experience of phenomena and material evidence, where rationality entails consistency over the entire belief-system. The systemic consistency-condition allows for post-hoc detection of falsity in the epistemic account of facts without having direct access or knowledge of all the relevant facts.
First-hand experience has the highest evidential value in relation to events and phenomena, but many important facts are not directly observable. For example, important scientific facts are not like self-evident phenomena but require complex technological mediation and specialist conceptual tools for ‘correct’ interpretation of the results according to the ruling theory. Modern scientific conclusions are inseparable from the evolving body of scientific and philosophical literature, and can only be meaningfully evaluated in that context. Prudence (as well as the first rule of fact-checking) demands that nothing in the literature should be taken as true in its own right, since the extent of scientific fraud, undeclared conflicts of interests, exaggerated claims, failures of peer review and honest errors is unknown. All references in the process of fact-checking must be fact-checked as well, with special regard for the human factor. This requires an immense amount of reading and academic-level analytical skills, and may therefore be prohibitively laborious even for professional fact-checkers.
Alternatively, fact-checking may proceed by verifying testimonies about the alleged facts by witnesses, authorities and experts. This can be reliably accomplished by in-depth interrogation of the relevant source with the aim of analysing its testimony for systemic coherence and logical continuity. Truth is systemically consistent; lies are not – they are logical conflicts or discontinuities in the system of relations. Genuine, phenomenal experience has practically infinite contextual richness and vague limits, and therefore can yield an unlimited amount of consistent information; made-up stories, on the other hand, are constructed from a limited range of well-defined conceptual ingredients. It is therefore possible to detect any lie, logical error, or fabricated story, given sufficient interaction with the source. Multiple verified accounts can then be correlated to obtain a more or less objective view of the state of knowledge of the relevant fact.
How to tell if someone is lying. Whenever I am suspicious about veracity of important information, usually some claim made by the government or a corporation, I try to contact the most relevant person or department and ask for clarification of one or two details of the claim. In case of a government department or public corporation it may take a few messages (I keep everything in writing) before the point of contact accepts that my inquiry will not be satisfied with the stock-standard PR blurb. By that time my query may be bumped up from the initial point of contact to someone higher in the corporate hierarchy. I continue politely asking specific questions and questioning the always incomplete answers. I don’t make any counter claims unless I can provide a citation. The goal of this exercise is not so much to get the truth and nothing but the truth (this never happens), but to keep the source talking. If a person is lying or makes unverified claims, then the more they talk the more false information they must keep track of in order not to contradict themselves. Be polite, professional and keep them talking. If talking continues and does not become circular (a typical defence of a liar which is also a kind of confession about dishonest motives) there is mathematical certainty that the contact will, at some point, contradict parts of their own account. Contradiction may of course happen naturally, as an honest error, but in such cases it can always be explained and corrected when challenged. A liar, on the other hand, will run into ever more contradictions while trying to explain prior contradictions. Truth is systemically consistent, lies are essentially not…
Polite interrogation is perfectly suited for fact-checking the fact-checkers. Posing as the ultimate arbiters of truth, institutional fact-checkers are ideally placed to act as political disinformation and opinion-formation (propaganda) agents. He who controls fact-checkers controls what counts as the truth. The social benefits of the fact-checking enterprise “must, at the very least, be weighed against the strong possibility that what passes for fact checking is actually just a veiled continuation of politics by means of journalism rather than being an independent, objective counterweight to political untruths.” (Uscinski 2015) The partisan nature of the dominant fact-checkers is undeniable, at least on political issues (Ostermeier 2011; Forbes; Daily Mail). “Much of what they decide to ‘fact-check’ is subjective at best. Nothing that can be pinned down with undisputed data.” (The Washington Times) Fallacious reasoning is a matter of policy: the absence of proof is taken as the proof of absence (PolitiFact). Findings of different fact-checkers poorly correlate with one another. (Marietta et al. 2015) It is therefore imperative to fact-check the fact-checkers before and above all else. We are told that fact-checking is built on trust (The Codex), but I contend that this is not genuine fact-checking. Diligent fact-checking is incompatible with trust, and therefore the only genuine fact-checking is when you do it yourself. This could well serve as the second law of fact-checking: no source shall be taken as prospectively truthful.
3916 views
” The goal of this exercise is not so much to get the truth and nothing but the truth (this never happens), but to keep the source talking. ”
Sounds good, doesn’t work if the interviewer has low social status and the liar has high social status. Try talking to a professor of physics about flaws in science. The professor will simply dismiss you as ignorant and unworthy to talk to him. People of high social status are adept at short-circuiting this type of “interrogation” based on their social status. Thanks for interesting ideas, and if posible I will write a nontrivial rebuttal about why I think you are wrong. I tried to hit the like button and it didn’t work for me.
Addendum to previous comment:
I should use a more nuanced argument: I have tried exactly what you describe many times, in various contexts, and I almost always have failed. If you have actually applied this to nontrivial problems and succeeded, then you should write a book teaching your method to the world, because you must be some kind of genius.
I agree that this process can be sabotaged by people, but that also justifies tentative dismissal of their claim. In the example of the physics professor, I can understand being somewhat dismissive if a student poses major objections to the established body of knowledge without actually reading up on the subject, in which case the professor is justified in merely directing the student to the relevant literature.
Yes, I have applied this method with great success a number of times. It works well when dealing with organisations or government departments who are legally obliged or otherwise committed to reasonably respond to public inqueries. I have, for example, led a vice-director of a governement department responsible for safety assessment and licensing of pesticides to admit that they simply collect the fee and sign of on any application that lands on their desk, as long as the company declares the product safety studies were conducted (by an ‘independent’ lab) and the paperwork is in order. In effect, the relevant authority admitted to having no first-hand knowledge of whether a product is safe or not, and allowed the industry to regulate itself, despite being charged with regulating the industry.
I have also found this approach most effective in dealing with police charges. I have for example, via polite questioning of procedural requirements, obtained an implicit admission that the Police failed to follow procedural requirements in proceedings against my friend. The
Police officer did not even realise that he did anything wrong. The case was dismissed on that basis Despite the admission of guilt.
It will not work every time, but in the right context it can be very fruitful.