Download Knowledge by Acquaintance and Knowledge by Description and more Exams Epistemology in PDF only on Docsity!
Objectual knowledge^1 Katalin Farkas in: Jonathan Knowles and Thomas Raleigh (eds.) New Essays on Acquaintance. Oxford University Press 2019. pp. 260- Abstract: It is commonly assumed that besides knowledge of facts or truths, there is also knowledge of things–for example, we say that we know people or know places. We could call this "objectual knowledge". In this paper, I raise doubts about the idea that there is a sui generis objectual knowledge that is distinct from knowledge of truths.
- Kinds of knowledge The Western philosophical tradition has focused on knowledge of truths as the central and paradigmatic case of knowledge. Here is a typical first sentence of an introduction to a book on epistemology: Epistemology is the theory of episteme, of knowledge. Ever since Plato it has been thought that one knows only if one’s belief hits the mark of truth and does so with adequate justification. (Bonjour and Sosa 2003, 1) Occasionally, before launching into the discussion of knowledge of truths, treatises on epistemology mention other kinds of knowledge. Again, the following characterization is typical: It is common in epistemology to distinguish among three kinds of knowledge. Thereʼs the kind of knowledge you have when it is truly said of you that you know how to do something—say, ride a bicycle. Thereʼs the kind of knowledge you have when it is truly said of you that you know a person—say, your best friend. And thereʼs the kind of knowledge you have when it is truly said of you that you know that some fact is true—say, that the Red Sox won the 2004 World Series. (Fantl 2016) I am going to call the three kinds of knowledge mentioned here practical , objectual, and factual or propositional knowledge. Since it has been in the center of attention, our understanding of factual knowledge is quite advanced. Factual knowledge, on virtually everyoneʼs opinion, is a specific kind of possession of a truth. A few further ideas are widely accepted as well (though each have a few dissenters). Possessing a truth means to have a mental attitude towards something that can be true or false, usually regarded to be a proposition. According to many, the mental attitude in question is a belief, but true belief is not sufficient for knowledge, there are further conditions: justification, or production by a reliable process, or by a cognitive ability or intellectual virtue, integration into an agentʼs cognitive character, a modal condition of “safety” or “sensitivity”, and so on. One function of these further conditions is to exclude a merely accidental or lucky possession of truth which is 1 Research on this paper was supported by project K-112542 of the Hungarian Scientific Research Fund, and I am grateful for the support received. Versions of the paper were given in the New Directions in the Study of the Mind project funded by the Templeton Foundation; at MIT; and at the University of Birmingham. I benefited greatly from the discussion after these talks. Iʼd like to thank Lukas Lewerentz for advice.
deemed to be incompatible with knowledge. On a somewhat different view, the relevant mental attitude towards a truth is not belief, but rather the mental attitude of knowing. Having this attitude excludes a merely lucky possession of truth by itself. So on a widely accepted conception, factual knowledge is at least a non-accidental possession of a truth (and possibly more). The nature of practical knowledge has been the subject of some debate. On the so-called intellectualist view, practical knowledge is a sub-category of factual knowledge (Stanley and Williamson 2001). According to the opposing, anti-intellectualist view–influentially defended by Gilbert Ryle (1949), among others–practical knowledge is not a species of factual knowledge, but rather some sort of ability or disposition to successfully perform a certain action. However, a plausible anti-intellectualist account of practical knowledge will make it clear why it is a species of knowledge. Presumably, defenders of the intellectualist view donʼt want to deny that that we can meaningfully talk about abilities to perform actions; what they deny is that these abilities constitute knowing how to do something. So the anti-intellectualist should be able to explain what is common to factual and practical knowledge that makes them both knowledge. One particular version goes as follows (Farkas 2018). Knowing how to G involves the ability to reliably succeed in a mentally guided execution of G -ing. Factual knowledge is one kind of cognitive achievement. It is an achievement partly because it involves success, where the particular success component of factual knowledge is having a true belief. The parallel success component of practical knowledge is a mentally guided successful execution of a certain type of action. In both cases, there is a mental attitude towards some content that is typical of the piece of knowledge in question. In neither case can the success be a matter of mere accident or luck. Some further conditions make sure that the success is non-lucky. So both factual and practical knowledge are at least non-lucky cognitive achievements (and possibly more). .
- Objectual knowledge–a third kind? On the conceptions sketched in the previous section, both factual and practical knowledge form a proper kind. Knowing different types of facts–eg. perceptual or mathematical, past or present, modal or actual–may have quite different conditions, but still, all instances of factual knowledge share a nature, in being some sort of non-accidental possession of a truth. Similarly, instances of practical knowledge–both on the intellectualist and anti-intellectualist account–share a certain nature. And since both kinds are properly regarded as knowledge, it looks like we have a genus – knowledge–with two species (or with a sub-species): factual and practical Let us turn now to objectual knowledge, which is the subject of this paper. I think it is fair to say that our understanding of objectual knowledge, as a kind, lags behind our understanding of factual and practical knowledge. There have been attempts to analyse some sub-categories of
truths, and the other to knowledge of things. He notes that the distinction roughly corresponds to the distinction between the German wissen and kennen , or the French savoir and connaître. The next chapter is devoted exclusively to knowledge of things, which, in turn, falls into two categories: knowledge by acquaintance and knowledge by description (also the title of that chapter). So on Russellʼs classification, knowledge by description is a subcategory of knowledge of things, rather than knowledge of truths. But itʼs clear from the examples he gives that much (or all) of knowledge by description is factual knowledge, rather than sui generis objectual knowledge. A more promising direction for our current purposes is to ask whether Russellʼs “knowledge by acquaintance” is sui generis objectual knowledge. Indeed it is quite common in contemporary discussions to equate knowledge by acquaintance with objectual knowledge (eg Zagzebski 1999, Martens 2010) Russell characterises the difference between knowledge by acquaintance and knowledge by description as a difference in their ground: knowledge by acquaintance is “logically independent from any knowledge of truths”, while knowledge by description “always involves (...) some knowledge of truths as its source and ground.” (Rusell 1912, 19) On Russellʼs view, acquaintance is a direct awareness of an object, without the mediation of inference or knowledge of truths. His primary example of acquaintance is our awareness of sense-data. It can be debated whether there is such a cognitive relation, and whether we are acquainted only with sense-data or possibly with other kinds of objects. But suppose there is such a relation; what is „ knowledge by acquaintance” as opposed to mere „acquaintance”? Or are these perhaps the same? In other words, is being acquainted with something a form of knowledge in itself? Some people deny this. John M. DePoe claims that direct acquaintance never suffices for knowledge; knowledge by acquaintance, like all knowledge, is knowledge of truths, and its distinguishing feature is that itʼs based on being acquainted with something (DePoe 2013). Richard Fumerton and Ali Hasan think that Russell probably „equivocates between the relation of acquaintance and the special kind of knowledge of truth (foundational knowledge) whose sole source is acquaintance” (Hasan and Fumerton 2014). To avoid confusion, they propose to restrict knowledge by acquaintance to foundational knowledge of truths. Knowledge by acquaintance, that is, knowledge of truths based on acquaintance is an interesting topic that raises a number of important questions, but I want to put it aside for the moment because itʼs not our central concern. My interest here is whether there is separate kind of knowledge which relates to things rather than to truths. Itʼs not clear to me whether Russell is after such a knowledge. Clarifying this issue would require a closer look at the relevant passages, something I have no space to do here. If acquaintance is the key to objectual knowledge, I propose to consult others on this issue who address this point much more clearly and explicitly than Russell does (see section 9 below).
Before we move on, let me note another unhelpful idea put forward by Russell. In the paper “Knowledge by Acquaintance and Knowledge by Description” (Russell 1910), Russell introduces the contrast between the two types of knowledge by talking of a case where “we know propositions about "the so-and-so" without knowing who or what the so-and-so is” (Russell 1910, 108). In these cases, he says, the subject is merely described. In the subsequent discussion, knowledge by (mere) description is again contrasted with knowing who or what someone or something is (Russell 1910, 113). It looks like Russell equates knowledge by acquaintance with cases of “knowing who” or “knowing what”. This isnʼt a course worth following. As we shall see below, for example “knowing who” is most plausibly a form of factual knowledge, and it clearly comes apart from ordinary cases of knowing someone. I know who Maryam Mirzakhani was (a mathematician, so far the only female winner of the Fields-medal), but I didnʼt know her. The usual sense of “knowing what” arguably also comes apart from Russellʼs special sense of knowing something by acquaintance. I could know what you are experiencing at the moment–the taste of Marmite–without being acquainted with the taste of Marmite. So I propose to steer clear of Russellʼs discussion of the matter.
- The language of knowledge ascriptions The following should be uncontroversial: when we gain knowledge about portions of the world, we enter into all sorts of interesting relationships to the things that are included in that portion. Facts have things as constituents, and for example when we have a perceptual knowledge of a fact, we commonly perceive the things that constitute that fact. But this observation does not require postulating a third kind of knowledge. It would require a significant further argument to show that knowledge of things is more than, or different from, knowing facts about them. So why think that there is objectual knowledge? An initial motivation may come from the linguistic constructions we use to attribute knowledge. Factual knowledge is often attributed by using that-clauses, as in (1) Isaac knows that apples fall to the ground. This may be contrasted with attributions where the verb „know” is combined with a noun phrase (or „NP” for short), as in the following examples: (2) René knows Marin. (3) Federico knows Rome. (4) Meno knows the way to Larissa.
differences. Reneʼs knowing Marin, or Federicoʼs knowing Rome are most plausibly understood as some sort of familiarity or acquaintance. But not all know-NP constructions express the kind of acquaintance that figures in these two sentences. For example, consider the sentence “Meno knows the way to Larissa”. On one understanding, this does mean that Meno is acquainted with the road: presumably, he has travelled to Larissa, he is familiar with the roadʼs various features. However, the same sentence can be perfectly well applied even if Meno has never been on the road, but merely knows which is the way to Larissa , under some appropriate description. Imagine that Meno is about to set off to Larissa for the first time in his life, after carefully studying and memorising the directions. If someone is wondering how Meno will get to Larissa, we can reassure her that heʼll be fine, he knows the way. The same ambiguity is present in many other noun phrase attributions. (5) could imply that Endre has been to the cafés in question, but it is also applicable if Endre has never been to Budapest, and just learned the names of the places from a guide book.^2 On one plausible proposal, the non-acquaintance reading of many of these sentences can be reformulated: Meno knows the way to Larissa just in case he knows which way leads to Larissa. This directs our attention to another group of attributions: know-wh attributions. These combine the verb “know” with a so-called “wh-clause” containing an implicit question, as in the following examples: (6) Cain knows where his brother is. (7) Eve knows which fruit is forbidden. According to the standard analysis, someone knows-wh only if she knows a/the proposition that answers the question implicit in the wh-clause. There may be additional conditions for sufficiency, but even if there are, know-wh is basically a type of propositional knowledge. Thus (7) is true because Eve knows that the apple from the tree of knowledge is forbidden , and this answers the question “Which fruit is forbidden?”^3 The proposal I just considered is that the non-acquaintance sense of know-NP attributions is equivalent to a know-wh attribution^4. Know-wh, in turn, is the same as (a possibly special kind of) propositional knowledge. Suppose that Meno knows that the E75 leads from Athens to Larissa; 2 In Hungarian, the non-acquaintance and the acquaintance senses of know-NP attributions are often translated by the two different verbs that correspond to “know” (“tud/ismer” as mentioned above; “Menón tudja az utat” / “Menón ismeri az utat”). As I understand, German is somewhat similar, at least for some objects, so “Menon weiss den Weg” and “Menon kenne den Weg” are both well-formed, and while the second implies acquaintance, the first doesnʼt. In French, “savoir” (the non-acquaintance senses of knowledge) does not naturally take noun-phrases (except for poetic contexts); in French, the natural translation of the non-acquaintance sense of a know-NP sentence is a know-wh sentence. 3 For a more detailed discussion of know-wh and for some proposals to refine the standard analysis, see Farkas 2016 and Farkas 2017. 4 See Brogaard 2009 for a discussion and defense of a similar view for “know-the” and “know-a” attributions.
since this answers the question “Which way leads to Larissa?”, Meno knows which way leads to Larissa is , which is equivalent to his knowing the way to Larissa. So the non-acquaintance sense of know-NP is equivalent to some sort of propositional (factual) knowledge. I suggested the outlines of an analysis for some knowledge-NP attributions via an analysis of know-wh attributions. It also makes sense to proceed the other way around. Many know-wh attributions are easily converted into a know-NP format, for example (6) can be converted to: (6ʼ) Cain knows the whereabouts of his brother. A plausible analysis for know-the sentences: S knows the F only if, for some x , she knows that x is the F. Other quantified phrases follow a similar pattern, and additional conditions may be required for sufficiency. Then we can proceed to give this analysis for the equivalent know-wh sentence. On either way of proceeding, we find that that the core of both knowledge-wh and a lot of knowledge- NP is knowledge of propositions. If the states attributed by many know-wh and know-NP sentences are in fact states of propositional knowledge, one may wonder why we have these other forms of attribution at all. There are good reasons: both forms have functions that know-that attributions cannot serve. First, one can attribute knowledge by using know-wh and know-NP even if she herself is ignorant of the matter. Someone who doesnʼt know the way to Larissa can still claim that Meno does. Know-wh and know-NP can also be used to attribute knowledge to a person over a changing subject matter, where factual attributions would require specifying many distinct propositions; for example, in saying that Kate always knows (what) the latest fashion (is). Both uses are made possible by the fact that neither know-wh nor know-NP constructions identify explicitly a propositional object in their content–even if they are in fact made true by knowing a proposition. To sum up this section: some cases of knowing the Fs or knowing some Fs are best understood as cases of know-wh, which, in turn, are best understood as some kind of propositional knowledge. What suffices for “knowing who” or “knowing which” is a highly context-dependent matter. For example, there doesnʼt seem to be a privileged list of propositions about a person whose knowledge would be either necessary or sufficient for knowing who that person is, in all contexts. But even if we cannot generally single out these propositions, in each context, knowing who someone is will simply consist in knowing some propositions. If all thing-related knowledge was like this, then there would be no sui generis objectual knowledge. So we need to search further.
- Knowledge in the biblical sense Within the category of knowledge of things, certain kinds of entities have received particular
One of the central cases of objectual knowledge is knowledge of persons; all discussions of this topic that I have seen include an example of knowing a person. (Witness the quote in the first section, and the quotes in section 9). What are the conditions for knowing someone? As a first approximation, we know people whom we have met (possibly a few times). Meeting someone involves mutual recognition of subjects as particular individuals. A similar thought is expressed by the Oxford English Dictionary which gives the primary meaning of “Know [with object]” as follows: “Have developed a relationship with (someone) through meeting and spending time with them; be familiar or friendly with.” (Stevenson 2010, 975) Philosophers will want to unpack these conditions further. Matthew Benton offers a plausible and carefully argued analysis of knowing people. Benton argues that the following is a necessary condition for “interpersonal knowledge” (knowledgei) ENCOUNTER: S knowsi R only if (i) S has had reciprocal causal contact with R, in which (ii) S treats R second personally, and (iii) R treats S second-personally. (Benton 2017, 822) Elaborating on this, Benton writes: A treatment by a subject S toward its recipient R is second-personal in virtue of S treating R as a subject (a “you”), where S offers some of S’s own thoughts, words, or emotions to R, and S is, or for the most part intends to be, attentive to R’s thoughts, words, or emotions. (Benton 2017, 822) This analysis accords with the Oxford English Dictionary definition. The condition of reciprocal causal contact is more flexible than the condition of meeting and spending time with someone; it allows for less direct forms of contact. Bentonʼs analysis is also much more informative on the kind of relationship that develops between people who know each other, by appealing to the idea of second-person treatment. On this conception, knowing persons is a symmetrical relationship: we know people who know us. It may be objected that we regularly talk about knowing people even if the knowledge is one-sided. Someone could claim “I know that woman, she lives next door” after spotting her on a photograph, even if they havenʼt met. As a turn of phrase, a political commentator could say “I know the president, she would never consent to this” even if they never met. In these cases, the knowledge in question plausibly consists in knowing relevant information about a person. I donʼt want to deny that on some occasions, this is how we talk. But I do think that the central and primary meaning of knowing someone is the one that implies personal interaction. In both of the cases just mentioned, it would make sense to ask: “I didnʼt realise you know her; when did you meet?”, and expect them to back down by clarifying that they donʼt know her personally, just know who she is, or what she is like.
Knowing someone produces propositional knowledge of the known person, and without some such knowledge gained, we would be reluctant to say that the subject knows the other (Crane 2012). But while knowing someone involves propositional knowledge, it has been argued that it is not reducible to propositional knowledge. This has been stated briefly for example by Ernest Sosa. The reason he gives for this irreducibility is that “(k)nowing someone or something, knowing some “object” in the broadest sense of this term, seems at least sometimes to require having had some special causal interaction with that “object.”” (Bonjour and Sosa 2003, 100). Benton expresses a similar idea: “I … locate the irreducibility of interpersonal knowing in the two-way interactions characteristic of two subjects treating each other as subjects”. (Benton 2017, 823) The idea needs elaboration. In a sense, some types of factual knowledge ‟require a special causal interaction” with an object. For example, it is a necessary condition for having perceptual knowledge that a subject perceives, or has perceived an object; otherwise the knowledge wouldnʼt be perceptual (that is, produced or justified by perception.) Perception, in turn, involves a certain causal interaction between an object and a subject. Seeing something involves an object visually stimulating the perceptual system of a perceiving subject, and visual stimulation is a causal process. That such a causal process is required by, or involved in, perceiving an object, can (and should) be admitted even by those who think that perception cannot be reductively analysed in causal terms. One could argue that the knowledge relation itself doesnʼt involve causation–itʼs only its production or justification that does. But since a certain type of production or justification is essential for this type of knowledge, there will be still some sense in which knowledge ‟requires” a causal interaction with its object. Another interesting comparison is with testimonial knowledge, which requires a certain kind of interaction with another subject. Perceptual knowledge and testimonial knowledge thus arguably “require” a specific kind of causal interaction with an object or with another subject, yet they are cases of propositional knowledge. Hence someone who wanted to argue for a reduction could push back in response to Sosaʼs and Bentonʼs observations, by suggesting that knowing a person is simply propositional knowledge that is produced through a certain type of causal interaction, or justified in virtue of standing in a causal relation to an object. In other words, it is structurally similar to knowledge by acquaintance, as discussed in the previous section. (On a certain notion of acquaintance, it would not be merely similar to, but also an actual case of, knowledge by acquaintance). I do not think this is quite right, but I wonʼt try to present a conclusive argument against such a reduction. Instead, I will split my argument into two strands. My overall claim is that there is no sui generis objectual knowledge which is both distinct from factual knowledge and also recognisably a member of the same genus “knowledge”. If you think that interpersonal knowledge is reducible to propositional knowledge, youʼll agree with the conclusion. If you think that interpersonal knowledge is not
and this explains its connection to knowledge in the paradigmatic sense.
- Knowing places, facts, rights, experiences In addition to knowing persons, knowing places is also frequently mentioned as an example of knowing things. Knowing places seems to share with knowing persons the feature that a special kind of causal interaction is required with the object. In order to know Budapest, one had to be there, for an extended times or several times, and plausibly, one also has to be able to get around at least to some extent. In this case, it is less clear than in the case of interpersonal knowledge that anything beyond factual and practical knowledge is required. As proposed before, in this case it is an additional condition that the factual knowledge has to be produced in a certain way, namely, through repeated personal interactions. Objectual knowledge attributions can take all sorts of objects, and in the case of some kind of objects, the knowledge is straightforwardly propositional, and there is no restriction on the production of knowledge. The most obvious case is knowledge of facts, or knowledge of the truth. (I have in mind attributions which would be translated by “connaître” and “kennen” and so on). Knowing a fact is factual knowledge. The same holds for some other kinds of objects; for example, knowing your rights, or knowing the rules, or knowing a theory, all plausibly consist in knowing some propositions, however that knowledge is produced. These are clearly quite different relations from the relation of interpersonal knowledge, and also different from knowing a place (which has to have a specific mode of production). Yet there is no obvious argument to exclude them from the instances of objectual knowledge. This suggests that the relation of “knowing an object” is different when it comes to different types of objects. Next I will consider yet another type of object where the relation is again arguably different. A frequently mentioned example of knowing things is knowledge of experiences. Sosa mentions knowing “the experience of a cold shower after a hard run” (Bonjour and Sosa 2003, 100), and the two philosophers considered in the next section talk about knowing “the agony of defeat” (Conee 1994, 14) and “the joy of victory” (Tye 2009, 96). To know an experience requires that one has had the experience. Unlike in the case of knowing people and knowing places, there is no requirement of repeated interactions. It is commonly assumed that once a person feels the agony of defeat, she will be at least in a good position to know the agony of defeat. However, there is something special in this case: namely, it is often claimed that the relation we have to features of experience when we have the experience has some features that deserve a separate discussion. This point leads us to the consideration of the philosophical notion of “acquaintance”. Most contemporary discussions trace back the notion to Russell, and as the definition in section 3 suggests, it is commonly assumed to be a direct (non-inferential) form of awareness of an object. It
is also often assumed that one can have this relation only to specific type of entities, primarily of a mental nature–mental facts or sense-data or features of conscious experiences. Russell thought that in addition to sense-data, one can also be acquainted with universals, and possibly with the self, but he categorically denied that we can be acquainted with physical objects or with persons as far as they appear to us physically. So knowing oneʼs best friend is certainly not a relation of acquaintance on Russellʼs definition. In the contemporary discussion, it is again frequently assumed that acquaintance is a special relationship we have to our conscious experiences; witness the following definition: Acquaintance is a unique epistemological relation that relates a person to her own phenomenally conscious states and processes directly, incorrigibly, and in a way that seems to reveal their essence. (Balog 2012) Balog makes some assumptions which may not be accepted by everyone: that acquaintance is incorrigible and reveals essences. But she shares the common view that the class of objects with which we can be acquainted is limited. On most accounts, acquaintance with mental items has an important role in producing knowledge about those items. But as we remarked earlier when discussing Russellʼs notion, this knowledge can be understood as propositional knowledge that is based on acquaintance, rather than objectual knowledge which is directed at an object rather than at a truth. Here we are particularly interested in the suggestion that acquaintance in itself constitutes (rather than merely produces or justifies ) objectual knowledge. The suggestion is that if one bears the relation of acquaintance to an object, then this is the same as knowing that object in the specific object-directed sense. I am going to consider two proposals along these lines in the next section.
- Acquaintance as knowledge of things Earl Conee and Michael Tye both suggest that acquaintance itself constitutes objectual knowledge. The context of both Coneeʼs and Tyeʼs discussion is the evaluation of the so-called “knowledge argument”. This argument is found already in Russell, and was then formulated independently by Howard Robinson (Robinson 1982) and Frank Jackson (1982). It concerns someone who has a certain type of conscious experience for the first time in their life. Tye and Conee both argue that the new knowledge gained on such an occasion–often characterized as knowing what itʼs like to have an experience–is non-propositional, objectual knowledge of (features of) the experience. In support of this position, they both argue that there is a general category of sui generis objectual knowledge. Conee claims that “(a)cquaintance constitutes a third category of knowledge,
different view, seeing something may count as an instance of acquaintance, and since we can see people, we can be acquainted with them. But even if this is right, acquaintance will still be different from interpersonal knowing. I see the person sitting at the next table in the cafe as I type these lines, but I donʼt know her. Acquaintance is also going to be different from other relations of knowing things. It is different from the relation we have to our rights when we know them. Suppose seeing is a type of acquaintance–seeing a place is not the same as knowing a place. I have seen a number of cities from the train on a journey two weeks ago, but I donʼt know these cities. So unless something else is said on the issue, we cannot assume that the same relation of acquaintance constitutes all the relations we bear to things when we are said to know them. There is another problem with the argument, and itʼs in the second step. If what is said in section 7 is right, then interpersonal knowledge does not belong to the genus “knowledge” whose nature is determined by factual knowledge as its central case. So we cannot use interpersonal knowing as one of the uncontroversial examples of, and model for, objectual knowledge, if our purpose is to give an account of knowledge that is the subject matter of epistemology. Initially, knowing persons may seem like a very plausible example of “a third category of knowledge, irreducible to factual knowledge or knowing how”, as Conee puts it. But I argued above that this picture is not right. Just what makes interpersonal knowing “irreducible to factual knowledge or knowing how”, will also plausibly prevent it from being a “third category of knowledge”, if “knowledge” is used univocally.
- Acquaintance as objectual knowledge, second try In the previous section, I considered an attempt to use the philosophical notion of acquaintance, understood as direct awareness of an object, to give an account of objectual knowledge. The idea was that acquaintance constitutes the ordinary relation of knowing things, and since these uncontroversially include cases of knowledge, acquaintance itself constitutes knowledge. I had two complaints about this idea: first, acquaintance in the philosophical sense is quite different for example from the relation of knowing persons; second, “connaître” may not be “knowledge” in the same sense as “savoir” is knowledge. But now note that the two problems could actually cancel out each other and hence open a new avenue for a theory of objectual knowledge making use of the notion of acquaintance. We could admit that acquaintance is quite different from the relation of interpersonal knowledge; but since interpersonal knowledge is not “knowledge” in the epistemic sense anyway, this isnʼt a big loss; in fact, it is an advantage. Next we could suggest that direct awareness of an object is relevantly similar to the central cases of knowledge, in being a non-accidental cognitive achievement. Objectual knowledge will then belong to the genus of knowledge, we just have to
keep in mind that ordinary attributions of know-NP, or uses of “connaître” and “kennen” most of the time do not ascribe “objectual knowledge” in the philosophical sense. Depending on the notion of acquaintance that is proposed to constitute objectual knowledge, the range of things that can be known in this sense will be more or less restricted. If we can be acquainted only with features of conscious experiences, then only those can be objectually known. If perception also provides acquaintance with the perceived objects, then we can objectually know everything we can perceive, but presumably not rights or rules. And objectually knowing for example a person will not be the same as knowing her. This, I believe, is the most promising avenue for developing an account of a sui generis objectual knowledge. But the account needs very significant further support. First, we would need an argument to show that knowledge constituted by acquaintance is needed in addition to (factual) knowledge based on acquaintance (ie. justified or produced by acquaintance). Many phenomena about acquaintance can be arguably explained simply by appealing to the idea of knowledge based on acquaintance. For example, it is often claimed that that we canʼt know what itʼs like to have a certain experience without having had the experience. One might think that this is explained by the fact that experiencing (which is the episode when one is acquainted with a phenomenal quality) is the same as what-itʼs-like knowledge. But thatʼs not the only possible explanation. Suppose that “what-itʼs-like” knowledge is always based on acquaintance; and the only way to become acquainted with a phenomenal quality is to experience it. This theory would also explain the claim. Second, even if acquaintance can be understood as a non-accidental cognitive achievement, this may not be sufficient for knowledge, even if it is necessary. For an example that is quite different from the case of acquaintance: perhaps certain products in the sub-personal processing of our perceptual system can be regarded as non-accidental cognitive achievements, yet at least on certain notions of knowledge, they do not qualify as knowledge. Or recall Wilfrid Sellarsʼs influential attack on the Myth of the Given–an argument to the effect that anything that “enters the space of reasons” has to have a propositional format. These discussions require significant further commitments about the nature of knowledge, and itʼs possible that all things considered, there will be a proper case for a separate category of objectual knowledge. In this paper, I tried to show that there is no easy route to establish the existence of a third kind of knowledge. The linguistic format of knowledge ascriptions has little significance; ordinary ascriptions of knowing things indicate a number of different relations to things, including, in a large part of the cases, the relation of having propositional knowledge about the object in question. Furthermore, we cannot assume without further ado that the relations expressed by “knowing things” are “knowledge” in the sense that concerns epistemology.
