That we are non-ideal agents is therefore highly relevant in this situation. We know that we may be susceptible to framing techniques that seek to discredit BLM, and we also know that should we consume media from across the political spectrum, we are highly likely to encounter those techniques. Following Worsnip’s suggested obligation would therefore pose a risk to BLM by making pernicious facts salient to a larger number of people.
Now, let us consider why Worsnip thinks we are obliged to diversify our media consumption. He claims that all publications, regardless of partisan affiliation, are illicitly influenced to a non-trivial degree. Illicit influences are those that lead a publication to report in a non-ideal way, leaving out important details or presenting the details in a way that is misleading, etc.
Having defined illicit influence, Worsnip’s argument is as follows:
- Worsnip’s Argument
 (Worsnip, 2019)
(4) seems an uncontentious claim about our psychology but it does highlight an important aspect of Worsnip’s theory, it is non-ideal. The distinction between ideal and non-ideal epistemic theories is in the agents that they apply to. Ideal epistemic theories consider ideal agents. Perhaps ideal agents would be able to adjust their beliefs appropriately once they recognise that their data set was incomplete, Worsnip’s argument would not apply to them. Non-ideal theories try to determine what course of action to recommend to non-ideal agents considering their failings. That Worsnip is acting in the non-ideal theory space will be relevant when we come to the objection and rebuttal. (6) is the premise which I will be indirectly contesting in the rest of this essay: in many cases, we are not obliged to diversify our sources. Given this creates a contradiction with the argument as stated, I suggest that (6) is misguided. Instead, we generally have a defeasible reason to avoid epistemically non-ideal beliefs by diversifying our media sources. With this in mind we can now turn to the objection.
We start with the premise that social movements such as BLM can be undermined by publications reporting only true claims. Banks’s 2018 work more fully explores the use of racial grammar, public memory, and framing in delegitimising BLM. Here, I want to focus on framing. In April 2015 Baltimore police killed Freddie Gray, a black man, while he was in their custody. This resulted in a largely peaceful protest, during which a small number of protestors rioted. Media coverage focussed heavily on the protest’s violent aspects, highlighting photos of the rioters over the peaceful protestors and keeping the debate focussed on the appropriateness of the rioting, rather than the virtues and necessity of BLM.
Written by James Shearer, University of St Andrews student
I have two issues with this response. First, the non-ideal aspect of the theory makes it difficult to see how actionable the advice to “diversify your media consumption, but only on some issues” really is. Sifting through publications looking for “safe” reports while avoiding the riskier ones is error prone for agents like us. Second is that I am not sure how many issues really are both important enough that we are obliged to be informed on them and free of any type of positive movement that can be undermined via pernicious reporting. It seems to me that for any given important issue, there is some side which I am under a moral obligation to support. For each, the argument regarding BLM will go through.
In an increasingly politicised society, previously mundane decisions about our daily lives can take on normative qualities. One such question is “what news media should we consume?”. Alex Worsnip suggests that we have an obligation to consume media from across the political landscape. This essay argues against this claim by showing that any obligation to diversify our media consumption in this way would face severe limitations. §2 will consider Worsnip’s argument. §3 will show why we are under no general obligation to diversify our media consumption. Finally, in §4 I consider and respond to potential responses to my position.
- We should expect all publications to be illicitly influenced to a non-trivial degree.
- Illicitly influenced publications are prone to omit important facts and stories.
- Reading only news sources from anyone side of the political spectrum will result in an incomplete picture of the evidence (From 1,2).
- We cannot rely on ourselves to adjust our beliefs to correct for the incompleteness of our evidence.
- Reading only news sources from any one side of the political spectrum will result in epistemically non-ideal beliefs (From 3,4).
- We are obliged to avoid having epistemically non-ideal beliefs.
Banks, Chloe (2018) Disciplining Black activism: post-racial rhetoric, public memory and decorum in news media framing of the Black Lives Matter movement, Continuum, 32:6, 709-720.
 (Ibid, Pg 243)
By focussing on these factors, publications make salient certain facts that can erroneously undermine our faith in legitimate social movements. Framing BLM in such a way as to link it with violent protest can delegitimise it when non-ideal agents are unable to adjust their beliefs appropriately in light of this framing. Given the politically charged nature of BLM, we can expect partisan media that is aligned against BLM to be illicitly influenced towards framing its reporting in this delegitimising fashion.
- Counter Case – Positive Social Movements
Finally, there is a response based on the distinction between epistemic and moral obligations. Epistemic obligations are derived from requirements of rationality, whereas moral obligations are based on requirements of morality. Worsnip is arguing for an epistemic obligation; it is an obligation based on the premise that we do better epistemically when we diversify our body of evidence. I, on the other hand, have been arguing that there may be countervailing moral obligations, but it is not obvious that we could not be obliged epistemically in one direction but morally in the other.
The claim is this; in many cases, we do something wrong when we follow Worsnip’s advice. We could not be obliged to do something that is often wrong so I conclude that we are not under a general obligation to diversify our media consumption. I want to examine a case where many people are going to be obliged to not diversify their sources on partisan lines. I will use Black Lives Matter (BLM) as an instance of a positive social movement, however the argument will generalise to any positive social movement that the reader prefers.
 Incidentally, this issue is why Worsnip cannot rely on his exception of publications that are “beyond the pale”. (Worsnip, 2019, Pg 258) Worsnip agrees that we are not obliged to consume publications that push immoral view points. My contention here is that unideal agents who try to diversify will inevitably do exactly that.
(1) is plausible; given the presumption that all publications have a political leaning, it seems that any given publication is going to have illicit, politically biased influences. We could dig into the differing degrees of influence, but I will grant Worsnip this claim for the purposes of this essay. (2) is a straightforward consequence of illicit influence.
- Response and Rebuttal
The second illicit influence more directly acts on the decision-making process regarding what facts to report. Consider the Ukraine example again, but this time the reporter actively believes that the donated weapons are killing civilians. However, because they still support NATO’s involvement, they do not want to do anything that will lead to public pressure against the intervention. This desire leads them not to report credible evidence of the deaths. Again, illicit influence is at play.
Worsnip claims that we are obliged to diversify the news media we consume such that we should read sources from across the political landscape, including those belonging to political positions that we disagree with. Before looking at how Worsnip structures and justifies his argument, I want to clarify the strength of his claim. There is a distinction between an obligation and a mere reason; Worsnip posits that we are under the former. I have a reason to attend office hours for my classes, it would improve my learning, but if I fail to do so I am not necessarily open to criticism. Perhaps I have a class scheduled at the same time; this would be a stronger reason to not attend office hours. However, if I fail to follow through on an obligation (say, the obligation not to kill) it seems that you automatically have licence to criticise me.
Worsnip considers two relevant ways that illicit influence might affect reporting. The first is through belief formation. Suppose a reporter believes that NATO should arm Ukraine. An uncomfortable consequence of arming the Ukrainian military is that some civilians will inevitably be killed by the donated arms. The desire to avoid acknowledging this tragic outcome might lead the reporter to wilfully ignore or discredit evidence that shows civilian deaths occurring because of these weapons. If they erroneously discredit the evidence or underreport the deaths, they do so under illicit influence.
Worsnip, Alex (2019). The Obligation to Diversify One’s Sources: Against Epistemic Partisanship in the Consumption of News Media. In Carl Fox & Joe Saunders (eds.), _Media Ethics: Free Speech and the Requirements of Democracy_. London: Routledge. pp. 240-264.
Before closing, I will consider two potential responses to my argument. First, we will look at a question regarding the extent of the limit I have argued for. We will then consider the importance of the distinction between epistemic and moral obligations.  (Banks, 2018)
Conclusion: We have an obligation to diversify our sources to be from across the political spectrum (From 5,6).
This article received an honourable mention in the graduate category of the 2023 National Oxford Uehiro Prize in Practical Ethics
Worsnip might agree that there are limits on the obligation to diversify our media consumption, but say that this is not the same as saying there is no obligation as such. It might be that we have a general obligation, but that it does not apply in the cases I have outlined. Perhaps I should not diversify my intake on social issues, but should do in other cases like on the economy.
If we think that supporting BLM is a moral obligation, then it would follow that we are obliged to avoid putting support for BLM at risk by making ourselves susceptible to discrediting techniques. This would constitute an obligation to avoid diversifying our media consumption in relation to BLM that would apply broadly to all people who are obliged to support BLM. Given this generalises out to all positive social movements that we are obliged to support, we consequently have a broad limit on Worsnip’s obligation. Now recall the distinction drawn between reason and obligation. That the obligation would be so often defeated suggests that we have a reason, not obligation, to diversify our media intake. I will not be attempting to offer a full account of the relation between epistemic and moral obligation here. I will instead settle with noting two problems that may emerge should Worsnip rely on this distinction in their defence. Firstly, Worsnip thinks that the obligation to diversify your sources is a case of moral and epistemic obligations “lining up”, but my objection shows that in many cases, this is not true. Second is that if moral and epistemic obligations do come apart as I have suggested, that would be a severe limit on said epistemic obligations. It seems to me that in these situations we do better to think of ourselves as having an epistemic reason that is defeated by our moral obligations. This would be a rejection of (6) in Worsnip’s argument.
This essay has argued that we have no obligation to diversify our media consumption. We began by looking at Worsnip’s argument to the contrary and his understanding of illicit influence. I then refuted Worsnip’s argument by looking at obligations to support positive social movements. If that argument goes through, then broad limits on the obligation will emerge. We would do better to take ourselves as having a mere reason to diversify our media consumption.
Obligations are not totally general, I might rightly kill an unprovoked attacker to save my own life. Such instances indicate the limit of an obligation. If an obligation can be shown to be so limited that we can freely break it in a wide range of cases, then talk of an obligation becomes less natural. Instead, it starts to look more like we just have a reason to act. I suggest that Worsnip’s obligation is better thought of as a reason to diversify our media consumption; one that will often be defeated.