Unwanted knowledge transmission

John Greco (2020) sets out to create a unified virtue-theoretic account of knowledge generation and transmission in his book The Transmission of Knowledge. One of the advantages of his view is that it can defend the achievement view from a known strand of counterexamples. To accomplish that, he relies on joint agency being essential to knowledge transmission. Joint agency can be characterized as a sort of interdependent and interactive cooperation between speaker and hearer that share an intention to transmit knowledge. This paper introduces the phenomenon of unwanted knowledge transmission which shows that joint agency, as Greco presents it, is not essential to knowledge transmission. Four types of unwanted knowledge transmissions will be introduced that pose threats to Greco’s account by showing that shared intention, a key characteristic of joint agency, is not present in all instances of knowledge transmission. Finally, some potential ways to defend Greco’s view will be considered and discussed.


Introduction
John Greco (2020), in his book The Transmission of Knowledge, argues for an antireductionist theory of knowledge transmission that sees knowledge transmission and knowledge generation as two distinct phenomena, where one cannot be reduced to the other and each requires a distinct approach. So, in Greco's view, there are two ways of coming to know.
Firstly, individuals can generate knowledge from their competent agency (Greco, 2020, p. 98). This competent agency is manifested by the individual when they competently use their abilities, such as their perception, to come to know something for themselves. As an example, imagine that you are standing in a field and a sheep walks B Ísak Andri Ólafsson i.olafsson.1@research.gla.ac.uk 1 Philosophy Department, University of Glasgow, Glasgow, UK up to you. Using your reliable faculties, you come to know that there is in fact a sheep standing before you based on your visual perception.
Secondly, individuals can receive transmitted knowledge by being a part of a competent joint agency (where both hearer and speaker are acting competently), where the true belief acquisition is attributable to the competent joint agency itself rather than the competent participation of the participating individuals (Greco, 2020, pp. 98-99).
So, knowledge generation is when one comes to know of their own accord, while knowledge transmission is when one comes to know from someone else without the usual epistemic burden that knowledge generation generally entails (Greco, 2020, p. 1). Epistemic burden here can be thought of as the work required to generate knowledge by one's own efforts using one's own cognitive and sensory abilities. Knowledge transmission, then, allows for a division of epistemic labour 1 in a way that knowledge generation does not (Greco, 2020, pp. 3-4).
According to Greco, knowledge transmission essentially involves telling, a speech act similar to assertion but with the clear purpose of sharing knowledge, 2 and it is a kind of joint agency, 3 in which individual agents act together. Joint agency can be characterized as "a network of shared intentions and common understanding between the participating actors, as well as specific kinds of interdependence." (Greco, 2020, p. 47). Furthermore, "trust is essential to joint agency in general, and joint agency is essential to knowledge transmission. It follows that trust is essential to one important way of coming to know." (Greco, 2020, p. 48).
In cases of knowledge, true belief is always attributable in some way to the competent cognitive agency of the knower (Greco, 2020, p. 99). Joint agency and joint achievement explain how the success can be credited to the hearer's competent agency even if that agency is not the most salient part of the success. When a success is directly attributable to the joint agency itself, the success is also indirectly attributable to the individuals that are a part of the joint agency to the extent they contributed to the joint agency in a competent way (Greco, 2020, p. 99).
A motivating factor for Greco here is to avoid a known strand of counterexamples to traditional virtue epistemological views, that conceive knowledge as an individual achievement of the knower, by showing how they fail to account for the inherent social dimensions of knowledge (Greco, 2020, p. 94). One prominent counterexample of this sort is Lackey's creditworthiness dilemma. 4 1 This is referring to Sandy Goldberg's (2011) thesis of the division of epistemic labour.
2 Christoph Kelp and Mona Simion (2022) have investigated the relation between knowledge sharing and assertion in their book Sharing Knowledge. In their view, the function of assertion is to share knowledge with others. 3 Abraham Sesshu Roth (2017) has written a helpful summary of shared agency. 4 For an understanding of Jennifer Lackey's creditworthiness dilemma, consider her case of the Chicago visitor: "Having just arrived at the train station in Chicago, Morris wishes to obtain directions to the Sears Tower. He looks around, approaches the first adult passerby that he sees, and asks how to get to his desired destination. The passerby, who happens to be a lifelong resident of Chicago and knows the city extraordinarily well, provides Morris with impeccable directions to the Sears Tower by telling him that it is located two blocks east of the train station. Morris unhesitatingly forms the corresponding true belief." (Lackey, 2009, p. 29). Morris seems to form a true belief because of the testimony of the speaker. This means that his competent agency is not the most salient reason for his true belief, which goes against the achievement view of virtue epistemology. This is problematic because if we allow Morris to rely on the Greco's account, if successful, can defend the achievement view from such counterexamples by stating that in cases of knowledge transmission, the hearer depends on the speaker as a cooperating agent in joint activity and the hearer's true belief is thus attributable to the competent joint agency of both speaker and hearer (Greco, 2020, p. 101). This preserves the advantages of the traditional achievement view, namely, that knowledge can still be considered intrinsically and finally valuable because it is an achievement, and knowledge is more valuable than mere true belief because we value achievements over mere lucky successes (Greco, 2020, p. 101). Greco's view just adds that knowledge is sometimes a joint achievement that is attributable to competent joint agency.
Joint agency thus plays a significant role in Greco's unified virtue-theoretic account of knowledge generation and transmission, an account he finds to be a "significant revision to traditional virtue epistemology" (Greco, 2020, p. 98).
Here is the plan. The first section will elaborate on Greco's view on knowledge transmissions and how it relates to joint agency. In the next section we will examine joint agency in more detail and highlight the necessary conditions for it. We will introduce the epistemological phenomenon of unwanted knowledge transmission (UKT) that is at odds with the essential characteristics of joint agency, and thus with Greco's account of knowledge transmission. For the time being, UKT can be defined as successful knowledge transmissions that are undesired by either speaker or hearer, e.g., when someone spoils the ending of a movie for you by telling you the plot twist.
Note that UKT cases are not meant to refute the claim that trust is essential to joint agency, but that joint agency, regardless of how it relates to trust, is not essential to knowledge transmission. If any of these cases successfully show that joint agency is not essential to knowledge transmissions, then Greco's view is compromised. We will then consider potential objections to the problems UKT introduce.

Shared intention and knowledge transmission
The speech act of telling 5 and the corresponding joint action are closely related. When knowledge transmission is successful, the speaker tells the hearer something, Footnote 4 continued speaker's abilities, as well as his own, then Gettier-type situations cannot be ruled out. To see why that is, consider a Gettier-type case where the environment is less suited for the formation of true beliefs such as a fake barn case, in which an epistemic agent sees a barn in the distance and forms the corresponding true belief. However, the barn in question is the only real barn among many barn façades. In this example, the agent's abilities and efforts are an important factor of how they came to know that there is a barn in the distance, but they are still lucky that their belief is true (Lackey, 2009, p. 33). Morris is in a relevantly analogous situation and thus vulnerable to these types of Gettier-situations; he was lucky that he asked a passerby that happened to know the city extraordinarily well. To formulate Lackey's dilemma in a succinct way, consider that in cases of knowledge, S's competent agency explains why S has a true belief and if S's competent agency must be the most salient reason for S's true belief, then many cases of testimonial knowledge are ruled out (such as Chicago visitor). However, if S's competent agency does not need to be the most salient reason for S's true belief, instead being just one important reason for it, then Gettier-type cases are allowed to go through. So, we either rule out cases like Chicago visitor or we allow Gettier-type cases, in either case the achievement view is in trouble. 5 Philosophers that have focused on the speech act of telling in instances of testimonial knowledge are, for example, Edward S. Hinchman (2005), Richard Moran (2005Moran ( , 2018, and Charlie Pelling (2014). manifesting their intention to share the knowledge with the hearer. Furthermore, the hearer understands and shares this intention, thus becoming a participant in the joint action of sharing knowledge (Greco, 2020, p. 49). For example, say you are visiting a friend that lives in a city you are unfamiliar with. You want to know where the nearest botanical garden is, so you ask your friend and they tell you where it is. In this example, you and your friend act together to ensure that the knowledge transmission goes smoothly. You ask your friend in a clear manner and they try to convey their knowledge to you in a competent manner. You are aware of the general content of what they are about to tell you and you anticipate their reply to help you on your way to the garden.
According to Greco, speech acts and other kinds of actions require that the world cooperates when trying to do something for it to achieve its aim, that is, trying to do X is not sufficient for X to happen; there must be cooperation involved (Greco, 2020, p. 49). The act of telling, as a species of assertion, follows general characteristics 6 of speech acts, and "is a social institution for the transmission of knowledge" (Greco, 2020, pp. 49-50). 7 The speech act of telling is the action involved in testimonial exchanges and it expresses a kind of interdependency between speaker and hearer, which is also an essential component of joint agency. Joint agency, as it appears in testimonial exchanges, can be "characterized by a special sort of cooperation between speaker and hearer" (Greco, 2020, p. 18) and it essentially involves trust relations between them (Greco, 2020, p. 18).
In any communicative exchange, the speaker intends to be understood by their hearer and the hearer understands that this is the intention of the speaker. Furthermore, the speaker and the hearer cooperate to achieve the intended result and they depend on each other to make the appropriate contributions to the exchange (Greco, 2020, p. 57). Greco emphasises that joint agency is more demanding than general communicative exchanges because knowledge transmissions involve a shared intention to impart relevant information (Greco, 2020, p. 57).
Think of two people walking to the same restaurant, but neither person knows where the other is headed (Greco, 2020, p. 54). In this scenario we see that there is no shared intention between them even though they both have an intention to walk to the restaurant, but if they were walking to the restaurant together 8 then they would share their intention of going to the restaurant, and they would both be aware that this intention is present (Greco, 2020, p. 54). This kind of teamwork, 9 that manifests as a 6 Hinchman (2005, p. 567) has defined telling along these lines: S tells A that p iff A recognizes that S, in asserting that p, intends: that A gain access to an epistemic reason to believe that p, that A recognize S's (2)-intention, and that A gain access to the epistemic reason to believe that p as a direct result of A's recognition of S's (2)-intention (Leonard, 2021). 7 Greco adopts this from Elizabeth Fricker's view on telling, where she states that "telling is a social institution for the spreading of knowledge" (Fricker, 2006, pp. 594-596). 8 To learn more about these kinds of cases, shared activity, and their characteristic features, see Michael E. Bratman (1992). For further discussion on walking together, see Margaret Gilbert (1990). 9 For a discussion on group competence and teamwork, see Ernest Sosa (2007, pp. 94-95). shared intention, is a hallmark of joint agency 10 and "[k]nowledge transmission is to be understood in terms of success due to the competent joint agency of speaker and hearer acting together" (Greco, 2020, p. 19). As shared intention is an integral part of joint agency, we can see that to effectively argue against Greco's view it is enough to show that there can be knowledge transmission without shared intention. 11 We can now present Greco's account of knowledge transmission.
Greco's account of knowledge transmission Knowledge that p is transmitted from a speaker S to a hearer H just in case S successfully tells H that p. And that happens just in case: (1) S knows that p.
(2) S asserts that p with the intention of sharing knowledge that p with H.
(3) H understands and shares S's intention. 12 (4) S and H act jointly so as to bring about their shared intention (i.e., so as to consummate the speech act in condition 2) (Greco, 2020, p. 57).
Greco's goal is to create a new unified virtue-theoretic account of knowledge generation and transmission and he relies heavily on this account of knowledge transmission to do so. We can derive a core thesis of joint activity as it relates to knowledge transmission using conditions (3) and (4) of Greco's account. Call this Greco's Shared Intention Principle. 10 Other characteristics of joint agency that are generally accepted in the literature are as follows; (1) it involves a shared intention between the joint agents, that is, they understand their action as something that they are doing together, (2) it involves planning on the part of the joint agents about how they are going to carry that action out together, (3) it is interactive, so the actions of one agent affect the actions of the other, (4) it is interdependent, so an individual agent cannot perform the joint action alone in the same way; each agent is doing a part of the joint action (Greco, 2020, pp. 55-56). 11 Many thanks to an anonymous referee whose comments prompted this clarification.
12 Some clarification is needed here to explain how the hearer can share the speaker's intention to tell the hearer that p when the hearer does not know the particular proposition the speaker intends to transmit. One way Greco could explain this is by saying that we, as epistemic agents, have a standing disposition of intention as communicators. Unfortunately, Greco does not say anything about the temporal duration of the intention. This is not problematic in cases where the speaker is responding to a specific inquiry (for example, you ask your friend if they are coming to your party and they give you an answer) (Greco, 2020, Case 4, p. 33), but it difficult to accept that third-grade students shared their teacher's intention to tell them that France is in Europe (Greco, 2020, Case 5, p. 33). Note that Greco addresses how a hearer can share a speaker's intention even when the hearer is not the one doing the telling, but he does not explain how they can share the specific intention of sharing knowledge that p. One attempt to address this is to reformulate Greco's account of knowledge transmission to have speakers and hearers intend to share knowledge in general, instead of some particular knowledge that p. However, as will be shown in the objections section of this paper (page 19), going this route makes it difficult for Greco's view to explain what makes knowledge transmission special compared to e.g. communicative exchanges. To clarify, imagine two people that only share the intention to walk to any restaurant together. Eventually they arrive at a restaurant that only one of them (or, in some cases, neither of them) wanted to go to. This would still be considered an achievement that could be credited to their joint agency just as if they had arrived at a restaurant that they both liked, which seems undesirable to me. This reformulated account is also susceptible to Lackey's dilemma as it relies on this weaker notion of shared intention to be the most salient reason for how a hearer comes to know a proposition from testimony. I am grateful to an anonymous referee for prompting further clarification on this point.

Greco's shared intention principle
A speaker transmits knowledge that p to the hearer only if the speaker intends to inform the hearer that p, the hearer understands and shares this intention, and the two successfully cooperate to execute their shared intention successfully.
UKT cases challenge Greco's account of knowledge transmission as they exhibit successful knowledge transmissions that do not follow the shared intention principle (conditions (3) and (4) of Greco's account of knowledge transmission), and thus cannot be representative of joint agency. We can spot a potential objection straight away that goes like this: The features of joint agency that UKT attacks might be characteristic of joint agency, but they are not essential to it. Greco specifically talks about characteristic features of joint agency instead of essential features to avoid disputes about which features are necessary for the definition of joint agency, as he is arguing that transmission shares all of the characteristic features of joint agency and that stands regardless of whether the features of joint agency are characteristic or essential (Greco, 2020, p. 55).
Even if this is the case, the joint agency features that Greco introduces 13 are the ones he uses to advance his view, and he argues that transmission shares all the characteristic features of joint agency regardless of whether they are essential or not. We see that even if Greco would be using a notion of joint agency that would lack features some would consider to be essential, then this less-than joint agency is still essential to his view, and the challenges that arise from UKT cases would remain the same. We can now introduce four distinct types of UKT that demonstrate how there can be knowledge transmission without shared intention or joint agency.

Unwanted knowledge transmission
UKT cases show that it is possible to transmit knowledge without shared intention 14 between speaker and hearer. The aim here is to demonstrate how Greco's account breaks down when facing this phenomenon.
We will present four types of cases that pose serious problems to Greco's shared intention principle. These cases all have in common that there is knowledge transmission 15 happening that is unwanted by either speaker or hearer (or both, in some cases). First, we will look at irrelevant knowledge transmissions, where the hearer is largely indifferent to the knowledge that is being transmitted. Then we will introduce unintended knowledge transmissions, where either hearer or speaker has no intention for knowledge transmission to take place. Next, we will look at resented knowledge transmissions, where the knowledge being transmitted is not only unintended but actively resented by the hearer. Finally, we will present achievement-depriving knowledge transmissions, which are resented because they can deprive someone of earning an achievement in a virtue-theoretical sense.

Irrelevant knowledge transmission
Irrelevant knowledge transmission is a type of UKT where a hearer comes to know something by transmission but is indifferent as to whether or not they came to know it. Irrelevant knowledge transmissions of this sort are prevalent in our day-to-day lives, but they are mostly harmless; they generally do not get in the way of us coming to know the things we want to know. Nevertheless, there is still something epistemically interesting about them; here is an example.
MARVEL FANATIC. One of your friends just keeps talking about the Marvel universe. This friend goes on and on about superheroes, the multiverse, infinity stones, and so on. After many years of close friendship, you realize you have slowly come to know extraordinarily many things about this Marvel universe. However, you have never been interested in knowing any of those things; you don't even like superheroes.
We see that the hearer does not share the speaker's intention of transmitting knowledge in the same way they would in regular cases where both speaker and hearer share an intention to transmit knowledge. 16 This innocuous example hints that Greco's demand for shared intention might become problematic for his view; at the very least, here, the intention that the speaker's knowledge be transmitted is not shared by speaker and hearer to the same degree. But what if the speaker or the hearer (or both) did not have any intention of partaking in knowledge transmission? To answer that, let us turn our attention towards unintended knowledge transmissions to see how knowledge transmissions can occur without apparent intention.

Unintended knowledge transmission
Here are three kinds of knowledge transmissions that are unintended. Eavesdropping The speaker does not intend for knowledge to be transmitted (but the hearer does).
Eavestalking 17 The hearer does not intend for knowledge to be transmitted (but the speaker does).
Overhearing Neither speaker nor hearer intend for knowledge to be transmitted.
For clarity, we will now present three corresponding examples that show how these distinct kinds of unintended transmissions appear to us. Consider this case of 16 Greco employs a notion of intention that gives rise to a generality problem. We could argue that the hearer intends de re to have a conversation, in the sense that the hearer intends to have a specific non-UKT conversation with someone without intending de dicto to partake in conversations that could contain UKT. Still, the problem here is that the intention, as an operator, is in a referentially opaque context, where the identity conditions for intentions would need to be laid out before we can commit to a specific reading of Greco's notion of intention. 17 The origin of the term eavesdropping comes from the obsolete noun eavesdrop, which is the ground on to which water drips from the eaves (and eavesdropping is thus when someone stands within the eavesdrop of a house, intending to listen in on a conversation inside the house). I named the corresponding speaker-intent scenario eavestalking, in which someone speaks within the eavesdrop of a house intending to be heard by the people inside the house. eavesdropping in which a speaker transmits knowledge to a hearer but the speaker had no intention of transmitting knowledge.
CHRISTMAS PRESENT. Parents are discussing their child's Christmas present. Unbeknownst to them, their curious child is eavesdropping on their conversation and comes to know what they will be getting for Christmas.
Eavesdropping cases are well known in the epistemic literature of assurance 18 and telling 19 . When someone eavesdrops on a conversation, they are not issued assurance because the speaker does not intend for the eavesdropping hearer to believe what they say, therefore failing to satisfy the conditions of telling as a speech act. Even though the intended speaker was issued assurance and the eavesdropping hearer was not, the assurance in question seems epistemically superfluous, that is, it does not seem to affect the epistemic status of the eavesdropping hearer's belief (Leonard, 2021). Now, even if CHRISTMAS PRESENT could not be said to be a case of telling as a speech act, we can still consider other cases that seem to satisfy the conditions of telling. Look at the following case of eavestalking, in which a speaker transmits knowledge to a hearer, but the hearer had no intention of partaking in knowledge transmission.
GIFT SUGGESTION. Drew is having some friends over for coffee while their spouse, Jordan, is working from home in the next room. Drew has their birthday coming up next week and is worried that Jordan doesn't know what birthday gift to get them. Drew proceeds to talk loudly about their birthday in the hope that Jordan might hear them. As it happens, Jordan can't help but hear what Drew said and comes to know that Drew would like a new coffee machine.
Finally, consider the following case of overhearing in which a speaker transmits knowledge to a hearer and both desire the knowledge transmission to take place, but neither intends for it to happen. DISTRACTED STUDENT. A philosophy professor wants to transmit knowledge to all their students. One student does not pay attention and does not come to know the things the professor said during a lecture. Afterwards, the student accidentally overhears the professor speaking to a colleague about the presentation they just delivered and comes to know the relevant material. This is akin to two friends walking side by side to a restaurant without either friend realizing that they are walking next to each other. It is strange to say that those friends are walking to the restaurant together in the same way as they would if they had planned to go together. In these three cases we see that there is no cooperation beyond some fundamental application of cognitive and sensory faculties. We also see that even though there is no competent joint activity taking place, and the requirements of Greco's shared intention principle are not met (as the speaker and hearer do not share their intention with each other), these cases still seem to portray testimonial knowledge exchanges. However, if the examples presented here are allowed as knowledge transmissions, then the problems that arise are significant and widespread, not just for Greco's view but numerous epistemological views that concern testimony and knowledge transmission.
One way to fight back against these unintended knowledge transmission cases is to say that they are not cases of knowledge transmission at all, but rather just typical examples of knowledge generation. Greco has said as much; communicative exchanges in general do not necessarily amount to knowledge transmission and testimony can sometimes generate knowledge instead of transmitting it (Greco, 2020, p. 26). However, as we will see, it is challenging to explain why these commonplace testimonial exchanges should not be considered knowledge transmissions, and even if one could do so convincingly, then Greco's theory would be severely limited in scope. It would cease to be a theory of paradigmatic cases of testimonial knowledge exchanges and could only be applied to a seemingly artificially constructed subset of some such cases.
Furthermore, even if we were to concede this line of argument completely, there are still other kinds of knowledge transmissions that pose greater problems to Greco's shared intention principle and thus his account of knowledge transmission. To see where we are headed, consider the following case of overhearing that shows a knowledge transmission that is resented by both speaker and hearer.
RUINED SURPRISE. Taylor is planning a surprise party for their spouse, Tracy. Tracy accidentally overhears Taylor talking on the phone and comes to know about the surprise party.
In this case, much like the DISTRACTED STUDENT case, neither speaker nor hearer is intending for the transmission to take place, but unlike DISTRACTED STU-DENT, neither speaker nor hearer want the knowledge transmission to take place. This kind of unintended and unwanted knowledge transmission demonstrates the futility of our epistemic situation; we can find ourselves on the receiving end of a knowledge transmission even when we explicitly intend not to.

Resented knowledge transmission
The third type of cases that can cause problems for Greco's shared intention principle involve the transmission of resented knowledge, that is, knowledge that the hearer does not intend to acquire and, if acquired, would be resented. Resented knowledge transmission cases are hard to categorize because they generally depend on individual preferences; some people would resent coming to know something while others would not mind.
These cases can be thought of in terms of counterfactual conditionals as the hearer is not in the epistemic position to know whether they would want to know whether p until they have already been transmitted knowledge that p or not p. Hearers in cases of resented knowledge transmissions might thus say "if I knew what you were going to tell me I would have asked you to keep silent", or, "if I knew how you would answer my question, I never would have asked in the first place". Here are four examples of resented knowledge transmissions that correspond to different emotions. 20 AFRAID. After your yearly check-up, the doctor tells you that you have an asymptomatic terminal illness and have only a few days left to live. You wanted to know whether you were healthy but after hearing the diagnosis you might have wanted to remain ignorant and live out the rest of your life without worry, but as soon as the doctor told you the diagnosis, the cat is out of the bag. This could be especially egregious when considering the (either real or imaginary) effect positive thinking can have in conjunction with treatment.
JEALOUS. You are at a restaurant with your partner celebrating your anniversary. Your partner then tells you how they used to go to this restaurant all the time with their previous partner and, in fact, they proposed to them at the very same table at which you are now sitting. Although this is not some horrific revelation, we can imagine some people would experience discomfort. One might say: "I did not want to hear this at the start of the date, it kind of killed the romantic atmosphere".
DISGUSTED. Someone tells you about Loa loa, a parasitic worm that travels under the skin of infected humans, across their eyes and into their lungs. This knowledge could prove useful to someone that is going to a rain forest in West Africa, but most people would rather go about their day without it. ENVIOUS. A friend tells you that a mutual acquaintance, a former business partner of yours that betrayed you, is doing exceptionally well. After hearing this you find yourself feeling unhappy for the rest of the day.
Note that whether or not these cases exemplify UKT depends on the different personality traits and emotional states of the people involved; some people would not be bothered at all by any of these cases while others would experience them negatively. Still, we can accept that there generally exists some knowledge for any individual that they would not want to know for the sake of their emotional well-being. 21 One line of argument that can be used to counter these examples is that people that experience negative feelings in these cases are emotionally immature, or at least less-than perfect epistemic agents, not using their cognitive faculties as well as they should have. Even if we concede that it is unreasonable to expect models to account for personal flaws of individuals, there is yet another kind of UKT, characterized by its achievement-depriving characteristics, that is still problematic for Greco's shared intention principle.
Footnote 20 continued alleviate these concerns, consider that even if the reason for not wanting to know something (and the reason one should first contemplate before transmitting knowledge that might have an adverse effect on the hearer) is not epistemically interesting, the process in which the knowledge gets transmitted is of great relevance. In sum, even if the reason for not wanting to know something is not epistemic in nature, the way in which one comes to know despite having an aversion to do so is epistemically intriguing. 21 Note that there is a difference between not wanting to know that p, and not wanting p to be true.

Achievement-depriving knowledge transmission
Achievement-depriving knowledge transmission (ADKT) is a type of UKT that can disrupt an achievement-earning attempt 22 of a hearer. Spoiler transmission cases are archetypal examples of ADKT; they incorporate knowledge that, if transmitted, would be resented by the hearer because it would spoil their attempt at achieving something. 23 Consider the following cases of ADKT.
CROSSWORD PUZZLE. You are solving a crossword puzzle and you are on the verge of completing it; you just need to find the last word. A friend of yours walks past you, sees that you are struggling, and proceeds to tell you the missing word before you can let them know you did not want any help.
MATH PROBLEM. You are solving a difficult math problem to prepare for a test. You are halfway through the calculations and trying to figure out how to proceed. You are confident you will eventually figure out what to do next, thereby expanding your knowledge of how to solve these types of math problems. Your spouse, who happens to be a mathematician, sees that you are working on a math problem and tells you what you need to do to solve the problem, not realizing you did not want any assistance.
MOVIE SPOILER. You are waiting in line to see the new mystery thriller movie when a stranger that just got out of the theatre walks up to you and tells you how the movie ends, including the plot twists that lead up to the shocking reveal.
These cases describe situations in which you want to earnestly try to find the answer yourself by competently using your own abilities, but someone spoils your achievement-earning attempt by telling you the answer. Moreover, if they intend to tell you the answer, there is nothing you can really do about it; you are not cooperating with the speaker (or, in some cases, the testimonial knowledge source) and yet you come to know the things you are being told.
In these cases, we find successful testimonial exchanges that transmit knowledge from a speaker to a hearer, but they do so without shared intention as it has been characterized. Specifically, we see that knowledge that p is transmitted from a speaker S to a hearer H, where S successfully tells H that p, S knows that p, S asserts that p with the intention of sharing knowledge that p with H, but H does not share S's intention even if H understands it, and they do not act jointly to bring about their shared intention (the intention of S to share knowledge that p with H). So, Greco's shared intention principle does not hold here and his reliance on joint agency is misguided.
One objection that immediately comes to mind is that speakers in cases like MOVIE SPOILER prove themselves to be untrustworthy by the testimonial act itself, as intentionally spoiling movies is surely not the mark of a trustworthy agent. If we assume that hearers try to cooperate with speakers, using their cognitive abilities to encourage knowledge transmission, then their competent agency would make them sensitive to this kind of untrustworthiness.
One way to address this objection is to argue that although the speaker in MOVIE SPOILER could be said to be morally untrustworthy there is nothing to indicate that they should be considered epistemically untrustworthy. It could even be argued that, because spoilers generally need to be true for them to have the desired spoiler-effect, hearers in cases such as MOVIE SPOILER should be more inclined to believe what they are being told when they are being told something they do not want to know. Furthermore, in CROSSWORD PUZZLE and MATH PROBLEM, where the speakers do not act with malicious intent, there is nothing to indicate that they are untrustworthy (epistemically or morally).

Objections
What are the potential ways to defend against the cases presented here? There seem to be at least two distinct paths available. First is to argue that UKT cases are not actually knowledge transmissions, and Greco's shared intention principle is thus not compromised as it does not need to account for those cases. Second is to argue that UKT cases are not only cases of knowledge transmission, but that they fit within Greco's account.
If we were to argue that there is no knowledge transmission happening in cases of UKT, how would we go about it? We could say that UKT cases should not be considered cases of knowledge transmission because they involve a hearer that is not cooperating, which means that the hearer does not share intentions with the speaker and is therefore not competently partaking in joint agency. Sure, in some testimonial exchanges the hearer is relieved of some of the epistemic burden that usually comes with acquiring knowledge on your own, but that is not always the case. 24 If we think about a police investigator questioning a potentially uncooperative witness, we can see that even if the witness is telling the truth, they are not just passing knowledge on to the investigator in the same way they would when they tell their child that there is milk in the refrigerator. 25 Greco does not find this worrying, as he does not hold the view that all testimonial exchanges must transmit knowledge, even in cases where the speaker knows, and not even necessarily in cases where the hearer comes to know from testimony of a knowing speaker (Greco, 2020, p. 5). The transmission thesis just claims that in some important type of testimonial cases, a speaker knows that p, reliably testifies that p, and a hearer comes to know that p because of the testimony of the speaker (Greco, 2012, p. 21). Greco could even say that UKT cases describe situations in which the expectation of cooperation is inappropriate. He claims that condition (4) of his account of knowledge transmission, that S and H act jointly so as to bring about their shared intention, cannot be satisfied by speakers and hearers in situations where the expectation of cooperation is inappropriate, 26 as the norms that govern cooperative testimonial exchanges are at odds with the norms that govern uncooperative testimonial exchanges (Greco, 2020, p. 59). So, condition (4) only requires that the speaker and hearer act appropriately to bring about their shared intention and this is not possible in uncooperative situations (Greco, 2020, p. 59).
However, this line of argument falls apart when we consider cases like CROSS-WORD PUZZLE, in which the speaker and hearer can hardly be said to be uncooperative beyond the fact that the hearer did not like what they heard, but this is wildly different from the cases that Greco has in mind here (e.g., the police investigator). If cases like CROSSWORD PUZZLE were not to be considered knowledge transmission, instead just being examples of communicative exchanges, it would be difficult to find exactly what makes knowledge transmission special. As previously discussed, Greco's theory would become severely limited in scope.
Furthermore, even in uncooperative contexts, many UKT cases, such as RUINED SURPRISE and MOVIE SPOILER, look like typical cases of knowledge transmission. This result should not come as a surprise; when we examine the nature of transmission, regardless of the context in which it is applied, we see that transmissions generally do not require trust or cooperation to be successful; a virus does not require cooperation to cause a pandemic.
More generally, we see that if Greco were to bite the bullet and claim that UKT cases are not cases of knowledge transmission because they do not adhere to his account in some way, he would simply be begging the question by claiming that certain kinds of knowledge transmissions should not be considered as such because they do not conform to the conditions of his account that details the necessary conditions of knowledge transmissions. Now, a different route to defend against the problems presented in this paper is to argue that UKT cases are compatible with Greco's account. One could say that the speaker and the hearer in UKT cases still share S's intention to tell H that p. Some actions can have both an individual action sense and a joint action sense, where the former refers to the actions of an individual and the latter the joint action of the participating actors. In acts of telling, like we have here, the joint action sense is the prevalent one, just like in speech acts of betting or promising. These unwanted knowledge cases are just exploiting this distinction by framing the knowledge transmission as being an individual action when it is not the case here. That is, even if the speaker intends to tell a hearer something, or more generally let them know about something, the hearer cannot share the speaker's intention as the hearer is not the one performing the action of telling or more generally, "letting know" (Greco, 2020, p. 60).
This objection is one Greco indirectly considers in his book. He says that knowledge transmission does involve the kind of shared intention that joint action implies, because the speech act of telling, much like the action of betting or promising, is ambiguous in nature; these actions can refer to the individual actions of a single actor, or they can refer to their part of a joint action, in which they share an intention with someone else to do something together (Greco, 2020, p. 60). 27 For example, when people make a bet there are two distinct actions in play, one is the individual action sense of trying to make a bet, 28 and then the joint action sense of actually making a bet with someone, that is, the bettor has not really made a bet with someone until the bet has been accepted, which would not make sense if the individual action sense of betting was the only one (Greco, 2020, p. 60). Furthermore, it seems like the joint action meaning of betting is the more commonplace one, which indicates that the individual actions in these cases are parasitic on the joint action meaning; if someone tries to make a bet and the other participant refuses the bet, the would-be bettor has not really made a bet (Greco, 2020, p. 60). Greco argues that it is not even necessary to claim that the joint action meaning is the prevalent one, all that is needed is that there "exists the joint action meaning of "telling," on the analogy with a joint action meaning of "betting" and "promising"." (Greco, 2020, p. 60).
However, there seems to be a stark difference between betting and telling; when a speaker intentionally spoils a movie for a hearer by telling them that Dumbledore dies it seems that the hearer "accepts" the bet automatically, as they come to know what they have been told and are aware of it when watching the movie for the first time, and it seems implausible to concede that they can refuse the knowledge transmission like they would refuse a bet (Greco, 2020, p. 60). What this means in sum is that unwanted knowledge can be successfully transmitted without the joint action meaning of telling, and even if we concede that this individual action meaning of telling is parasitic on the more common joint action meaning, it still allows for the unwanted knowledge transmission to go through.
Another way to argue that UKT is compatible with Greco's view is to say that there is actually some sort of cooperation happening. One way to formulate this response is to say something along these lines: Knowledge transmission, regardless of whether the knowledge is wanted or not, involves a sort of fundamental cooperation, where both speaker and hearer are competent agents that are competently using their cognitive abilities to have a successful testimonial exchange that transmits knowledge. If this is true, then UKT cases adhere nicely to Greco's account.
We can respond to this objection by pointing out that this sort of fundamental cooperation would not be the primary factor in the hearer successfully acquiring knowledge. Even if the speaker and hearer would be indirectly credited with success by individually applying their cognitive abilities to competently cooperate, the success of the 27 He takes an example of Brady throwing a ball into a practice net and how it is different from Brady throwing a ball to his teammate, Gronkowski. If the teammate does not catch the ball, then Brady merely tried to pass the ball, but was unsuccessful in his attempt. (Greco, 2020, p. 60). If Brady is to complete the pass, Gronkowski must catch it. We can respond by saying that this is not accurately reflecting what is really happening in cases of telling as a joint action. A more fitting example would be a case wherein Brady throws the ball as forcefully as he can towards Gronkowski's head with the aim of passing the ball to him (for him to pass the ball successfully it is enough that Gronkowski ends up with the ball in his hands as a direct result of Brady throwing it to him), Gronkowski grabs the ball instinctively so as to not get hurt. Brady has completed the pass, but Gronkowski did not share Brady's intention, and he would not be considered as a participant in a joint activity. 28 An individual action sense of making a bet is, for example, when someone says: "I bet I can fix this." transmission cannot be credited to their joint competent agency as the necessary conditions for joint agency are not met. In other words, as the hearer does not share the speaker's intention to share knowledge that p, they cannot act jointly to bring it about. 29 We can modify condition (2) of Greco's account of knowledge transmission (that the speaker asserts that p with the intention of sharing knowledge that p with the hearer) in a way that allows Greco's shared intention principle (that state that the hearer understands and shares the speaker's intention and they act jointly so as to bring about their shared intention) to hold. A modified condition (2)* states that the speaker asserts that p with the intention to share knowledge with the hearer. Note that the speaker in (2)* does not intend to share knowledge that p with the hearer, just knowledge in general. This way, conditions (3) and (4) hold, as the hearer understands and shares the speaker's intention to assert some knowledge and they jointly act to bring about their shared intention of the speaker sharing knowledge with the hearer.
If we were to accept this modified version of Greco's account of knowledge transmission, then it would be reduced in a way that all that could be said of testimonial knowledge exchanges and joint agency is that the relevant intention is just to share knowledge in the most general sense, with no regards to the content of testimony. 30 Furthermore, this arrangement fails when we come to know something we did not want to know, as there is no standard mechanism in place for us to unlearn what we have come to know. Even worse, we cannot reliably filter out unwanted knowledge in advance without knowing the contents of said knowledge; the more you inquire about the content of what is about to be said the closer you are to knowing it.
If joint agency were to be diluted in this manner to account for the cases presented in this paper, then joint agency would overgeneralise to cases of knowledge transmissions that are obviously not consensual. 31

Concluding remarks
The principal aim here has been to show that it is possible to transmit knowledge without shared intentions and show that Greco's account of knowledge transmission, which depends on these shared intentions through joint agency, is in trouble. Those who advocate for the sort of anti-reductionist knowledge transmission view that relies 29 Note that joint causation does not imply joint agency. So even if two individuals both have an intention to walk to the same restaurant, they are not partaking in joint agency unless they are intending to walk to the restaurant together. 30 Note that this would result in Greco's transmission account being broadened in a way that would make it difficult to grasp what exactly is special about knowledge transmission. That is, if shared intention is modified to be inclusive enough to accommodate the idea that all communicative exchanges are considered joint activity, then, for one thing, at least some testimonial Gettier cases will be considered creditable achievements. I am grateful to an anonymous referee for prompting further clarification on this point. 31 Note that this weakened form of joint agency also looks incompatible with a datum we find in the collective intentionality literature, namely, that praise and blame attributions can be appropriated to groups by shared agency. When shared agency has been stretched this thin it becomes unclear how group-level praise and blame would be appropriated as opposed to just individual-level praise and blame, as it is generally viewed as a function of cooperative interaction between the members of the group. on shared intentions, like the view Greco presents, find themselves in a dilemma when they are put up against UKT cases. Either they concede that joint agency is not a necessary condition for knowledge transmissions, or they insist that it is present in all cases of knowledge transmissions, including UKT cases.
If they go for the first horn of the dilemma and say that UKT are not cases of knowledge transmission, then many paradigmatic examples of knowledge transmissions would not count as such, which in turn makes Greco's view only applicable to a seemingly artificially constructed subset of some such cases. This means that there will be many cases of knowledge transmission that fall prey to Lackey's dilemma, as there is no competent joint agency that can be credited for the true belief of the hearer.
If they go for the second horn and argue that UKT cases are cases of knowledge transmission, then they would be forced to accept a diminished form of joint agency that would simply be an ever-present by-product of testimonial exchanges. The triviality of this form of joint agency would make it difficult to discern what exactly makes knowledge transmission special and this account would overgeneralise to cases of knowledge transmissions that are obviously not cooperative and do not follow Greco's shared intention principle. This weaker joint agency cannot explain how a success can be credited to the hearer's competent agency even if that agency is not the most salient part of the success, as the success cannot be directly attributable to this trivial version of joint agency.
The ramification of the dilemma introduced here is that competent joint agency is either not present in many typical cases of testimonial knowledge exchanges, which makes those cases susceptible to Lackey's dilemma, or it is present in a diminished form that cannot produce the sort of shared credit that is needed to defend against Lackey's dilemma. 32 Funding The author has no relevant financial or non-financial interest to disclose and did not receive support from any organization for the submitted work.

Conflict of interest
The author has no conflicts of interest to declare that are relevant to the content of this article.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/ by/4.0/.