Dentith, Matthew. "When Inferring to a Conspiracy might be the Best Explanation". Social Epistemology, Vol.30, No.6 (2016): 572-591.

Dentith, Matthew. "When Inferring to a Conspiracy might be the Best Explanation". Social Epistemology, Vol.30, No.6 (2016): 572-591.

• Source mine of philosophy of conspiracy theories on page 572.
• The main course which most philosophers take in criticizing conspiracy theories is that they are considered unlikely, in particular because conspiracy theorists propose that much of the historical narrative is dominated by large, successive conspiracy theories (573).
• Pete Mandik asserts that although they may be unlike, the causal narrative provided by conspiracy theories is more likely than coincidence or random chance in each individual case. Therefore, at first glance, believing a conspiracy theory is the more reasonable and likely option (574).
• Conspiracy theories may be very unlikely given a full range of evidence surrounding the events, however they provide the most satisfy explanation of events. In this sense, they are some of the least likely explanations, but also the 'best' explanations (574).
• Belief in conspiracy theories is also generally characterized by a refusal to accept official or conflicting information. Conspiracy theorists are actually more interested in criticizing official accounts than advancing particular alternatives (574-575).
• The claim, upon which most philosophers are willing to dismiss conspiracy theories, that conspiracy theories are very unlikely is not based on substantial evidence. American politics has witnessed a number of important conspiracies, including the conduct of politicians in the MacCarthy era, and the Gulf of Tonkin incident. Governments and companies have historically been very skilled in engaging in conspiratorial behavior (576-577).
• At its most basic level, a conspiracy theory must claim that actors intentionally arranged an event, that they have attempted to hide their role in this event, and that this event served a goal of the actors. Many real-life events fall into this category, including things like the Iran-Contra affair or the Watergate scandal (577-578).
• Pete Mandik and other philosophers claim that these later real conspiracies do not count as conspiracy theories because they were not kept secret. This is nonsense, this this would require real events to be kept perfectly secret to count as conspiracy theories, meaning that no conspiracy theories could ever be proved or known (577-578). 
• The lines are further blurred by the claims of Dr. Mandik and others that official account of events are not conspiracy theories since they are official. This makes sense -- few would call claiming that 9/11 was carried out by Al-Qaeda a conspiracy theory -- but raises questions about what makes conspiracy theories different from theories about conspiracies (578).
• Some scholars, such as David Cody, have provided alternatives to this view, requiring that conspiracy theories exist in opposition to the official version of events. Even with this distinction, however, Dr. Cody notes that many official theories are unlikely, conspiratorial, or fraudulent (579-580).
• Epistemological philosophers have generally considered two positions towards conspiracy theories: generalist and particularist. Generalists believe that all conspiracy theories can be judged erroneous without examination of specific claims because they are conspiracy theories (581). Particularists argue that conspiracy theories needed to be judged on individual merits like all other theories (582).
• Conspiracy theories, like other assertions, are considered on the basis of their posterior probability, extent to which evidence makes the theory likely; their prior probability, the initial likelihood of the explanation without evidence; and the relative probability, compared to competing hypotheses (583).
• These different forms of probability all affect the level of evidence needed to claim a conspiracy theory is true or false. If conspiracies are generally uncommon, then the prior probability is low and a high level of posterior probability is required. The strength of the alternative explanations also affects these factors (585-586).
• Conspiracy theories should be discussed and considered on their own merits, because they are not necessarily less likely than other explanations (587-588).