Brentano and the Medieval Distinction Between First and Second Intentions

Brentano’s account of intentionality has often been traced back to its scholastic sources. This is justified by his claim that objects of thought have a specific mode of being—namely, “intentional inexistence” (intentionale Inexistenz)—and that mental acts have an “intentional relation” (intentionale Beziehung) to these objects. These technical terms in Brentano do indeed recall the medieval notions of esse intentionale, which is a mode of being, and of intentio, which is a “tending towards” (tendere in) of mental acts. However, within the lexical family of intentio there is another distinction that plays an important role in medieval philosophy—namely, the distinction between first and second intentions (intentio prima and intentio secunda), which are, roughly speaking, concepts of things and concepts of concepts respectively. What is less well-known is that Brentano explicitly borrowed this distinction as well, and used it in his account of intentionality. This paper explores this little-known chapter in the scholastic-Austrian history of intentionality by evaluating both the historical accuracy and the philosophical significance of Brentano’s borrowing of the scholastic distinction between first and second intentions.


Introduction
Brentano had a groundbreaking approach to the philosophy of mind. He proposed to study our mental life from the first-person point of view with the help of concepts which are all abstracted from what is given to us in inner consciousness, and without the admixture of any physiological notions. He called this approach "descriptive psychology" or "descriptive phenomenology" (see Brentano 1982 and forthcoming). This famously influenced his student Husserl, and through him many other philosophers in the twentieth century. While making groundbreaking investigations, Brentano nonetheless also relied heavily on earlier intellectual traditions, especially Aristotelian-scholastic philosophy, which he knew very well: he wrote his dissertation (Brentano 1862) and Habilitation (Brentano 1867) on Aristotle and was a priest for a time. Brentano is surely the perfect author to choose in order to find connections between the debates about philosophy of mind in the Middle Ages and those in phenomenology, and in contemporary discussions as well, since current studies on the mind are still significantly shaped by those which took place at the turn of the twentieth century (due to Brentano's influence, among others). In short, studying Brentano's philosophy of mind is relevant from both a historical and a theoretical point of view.
Famously, Brentano himself connected his work to scholastic thought regarding a central notion of his philosophy of mind, namely, that of intentionality. He claimed that all our mental acts have an "intentional relation" (intentionale Beziehung) to an immanent object (i.e., an object in the mind) that has "intentional inexistence" (intentionale Inexistenz), and held that these ideas came from the scholastics. They do indeed recall certain medieval notions, which are also expressed with words from the lexical family of the Latin intentio, in particular the concept of esse intentionale, which is the specific mode of being that objects have when they are in the mind. However, there is another, less wellknown link between Brentano's account of intentionality and medieval theories of "intentions". It is based on an important distinction used in medieval philosophy, namely, the distinction between first and second intentions (intentio prima and intentio secunda). In fact, Brentano explicitly mentions this distinction in his account of intentionality; but since he makes the point only in unpublished texts, it has been missed by his readers.
The study of Brentano's borrowing must be made carefully, since the distinction between first and second intentions was understood in various ways in the Middle Ages. Now, since Brentano himself does not mention any predecessor, the identification of his possible sources requires some historical investigations. This in turn must be preceded by an exact understanding of the way Brentano himself uses the distinction. As I will show, Brentano holds that a first intention is what is thought about by our first-order (presumably conceptual) intentionality, while a second intention is the content of our (conceptual) mental acts and is given to us only in reflexive, or second-order intentionality. 1 Brentano helps himself to this distinction to address a crucial theoretical issue: the distinction between first and second intentions is supposed to help him to solve epistemological problems related to his (early) account of intentionality, in which intentionality is described as a relation directed at an immanent object, an account which seems to rule out our having cognitive access to extramental reality.
In the first part of the paper, I will sketch out some basic elements of Brentano's theory of intentionality. In the second part, I will reconstruct Brentano's use of the distinction between first and second intentions. In the third part, I will evaluate Brentano's use of this distinction, with respect not only to its faithfulness to scholastic sources, but also to its philosophical significance, thus entering into a detailed critical engagement with Brentano.

An Outline of Brentano's Theory of Intentionality (and Its Medieval Sources)
One central goal of Brentano's Psychologie is to identify the "mark of the mental", that is, the feature which distinguishes psychic phenomena from physical phenomena. Brentano aims to identify this feature in order to sharply delimit the investigations of psychology and distinguish it from other sciences. His solution explicitly relies on the scholastic philosophers of the Middle Ages: he holds that the mark of the mental is a specific directedness towards a "content" (Inhalt), or "immanent objectuality" (immanente Gegenständlichkeit), which has a specific mode of being, namely, that of "intentional inexistence" (intentionale Inexistenz). Importantly, the prefix 'in-' does not refer to a negation, but to an inclusion, meaning that the object is within the mind, and not outside it (Brentano 1924, pp. 124-125 andChrudzimski 2001). 2 These claims are usually summarized as follows: for Brentano, the mark of the mental is intentionality, and this notion is medieval. However, in the passage in question, Brentano speaks not of "intentionality", but of "intentional inexistence". So, exactly which element(s) in his account is (are) related to the scholastics? 3 If one were to use lexicography as a preliminary guide, there is a great variety of terms which are linguistically cognate with the Latin intentio and are used in scholastic philosophy. So, again, which of the scholastic senses is (or are) used by Brentano? In the Middle Ages, in discussions in the philosophy of mind, one finds the word intentio and cognates in the description of mental acts. There is the famous idea-found initially in Augustine, 4 but also defended by Thomas Aquinas 5 -that every mental act has a specific sort of directionality, an "intention" (intentio) or a "tending towards" (tendere in) an object. Though this feature seems to be active for these authors, this does not mean that only acts of the will have such an intention: even perception is equipped with an intention directed at its object.
However, the word and its cognates are also used to describe mental contents. There is one sense of intentio which is that of 'meaning' or 'concept', which originates in Avicenna's notion of maʿnā, primarily in a passage where he describes the grasping by a sheep of the hostility of a wolf. 6 This notion is further distinguished into first and second intentions (intentio prima and intentio secunda): roughly speaking, concepts of things, such as the concept horse, and concepts of concepts, such as the concept being predicable from a plurality. This distinction also has Avicennian origins. 7 Interestingly, in Hervaeus Natalis, 8 one finds the technical term intentionalitas, by which Hervaeus means a relation which goes from the object to the act and is the feature or property that makes the object an "intention" (just as whiteness makes something a white thing). 9 Another important case is the thesis that mental contents have a specific mode of being, namely, esse intentionale, often also called 1 The content is thus an entity distinct from that which is thought about by first-order intentionality. More on this in Sect. 3. 2 For a different reading, according to which immanent objects, despite their name, are not in the mind according to Brentano, see Antonelli (2001) and Sauer (2006). 3 For a good overview of theories of intentionality in the Middle Ages, and a comparison with Brentano's views, see, among others, Perler (2002). 4 See, e.g., Augustine, De Trinitate 11.2.2 (Augustine 1968, p. 334, among others). 5 Thomas Aquinas,De veritate,q. 13,a. 3,sol. (ed. Leon. 22.2.2,p. 424). References to Thomas Aquinas (ed. Leon.) are from Thomas Aquinas (1882-). 6 See al-Shifāʾ, al-Nafs 1.5 (Avicenna 1959, p. 43 1 3 esse obiective. This is a very common idea in the Middle Ages, which is found in Aquinas among many others. 10 So, what about Brentano? In the Psychologie (Brentano 1924, pp. 124-125), he clearly is using the vocabulary of intentio (in German: Intention) to describe the content side of intentionality, since he equates "content" (Inhalt) and "immanent objectuality" (immanente Gegenständlichkeit), and claims that this object (or objectuality) has "intentional inexistence" (intentionale Inexistenz). Although he speaks there of the mind's directedness, he does not label this directedness "intentional". Later, however, he also applies the vocabulary of intentio to the act side: he speaks of the directedness of the act towards an object in terms of an "intentional relation" (intentionale Beziehung) (Brentano 1982, p. 21), which he also calls "intentionality" (Intentionalität) (see ms. EL 81, n. 13508, quoted in Rollinger 2010, p. 24, n. 50). Thus, Brentano's philosophy of mind, and its central notion of intentionality, are indeed related to scholastic philosophy, terminologically as well as theoretically: not only does Brentano use intentio and its cognates to label mental contents, which he treats as objects with a specific mode of being, that is, as "immanent objects" with "intentional inexistence", he also uses the vocabulary of intentio to refer to the directedness of the mental, just as Augustine and Aquinas did.
However, Brentano's account of intentionality leads to an epistemological problem: we do not think about (or at least not only about) objects within our mind, but about things in extramental reality. So how is a philosophy of mind with immanent objects to be combined with an epistemology in which our cognitive capacities are directed towards the things themselves? I will argue that it is precisely in order to solve these issues, at least with respect to our conceptual presentations, that Brentano borrows another famous scholastic distinction, which, as mentioned above, is lexically related to intentio-namely, that between first and second intentions. This distinction will help him, or so he thinks, to distinguish between the directedness of the mind towards immanent objects and its directedness towards extramental reality. Thus, the distinction plays an important strategic role in Brentano's overall theory of intentionality.

The Textual Occurrences of the Distinction
The textual sources where Brentano develops his views about first and second intentions are rather meagre. To my knowledge, his only uses of this scholastic distinction are found in an unpublished manuscript, which is an appendix to ms. Ps 76, which contains Brentano's lectures on descriptive psychology given in Vienna around 1887/1888. However, as I would like to show, despite its few occurrences, this distinction plays an important role in Brentano's overall account of intentionality. 11 The text in which the distinction appears is dated 20 April 1875, and it is described by the editor of the Nachlass, Franziska Mayer-Hillebrand, 12 as a "fragment from a psychology lecture" (Fragment aus einem Psychologie-Kolleg) (see Brentano,Ps 76,. It consists of brief notes, mostly lists of distinctions between various central philosophical notions. The distinction between first and second intentions is found in lists about mental and linguistic representations and their objects. As indicated above, the usual view among medieval philosophers was that intentions are concepts, and first intentions are concepts of things while second intentions are concepts of concepts. As I would like to show, however, Brentano's account seems to differ from this standard medieval understanding.
The first occurrence of the distinction is the following (Brentano,Ps 76,: Names Presentation Content of presentation matter of presentation Object of presentation. Named. Presented real identity of the presented per se per accidens conceptual identity O ne content of presentation often many presented things (with the universal) T he content determines the presented often indeterminately presented = prima intentio Content = secunda intentio Man is a name Man is a species A man is a mortal animal (here the article) E ach presentation has a content of presentation, but not every one has an object: Without entering into too many details, a brief description of the main characters of the story will help to guide the reader: a presentation, in Brentano, is a type of mental act in which an object merely appears to us without any judicative or emotional attitude towards it, e.g., my visual experience of a horse, or my conceptual thought of horse; the content or matter is-so I will argue at least-what Brentano elsewhere calls the "immanent object", and is a sort of mediator or mental "image" (as a text quoted below says); and what he here labels the "object of presentation", also called "the presented", is a mind-independent thing which (possibly) corresponds to a presentation.
Note that the manuscript says that the content is (often) indeterminate, in contrast to the object, which is determinate, and so the content (often) determines (that is, points towards) the object indeterminately. What Brentano is referring to here are universal thoughts. In these cases, the content is indeed indeterminate-that is, it is lacking at least an individuating feature (e.g., here and now), and possibly also further specifications (e.g., thoroughbred), and is thus universal (e.g., horse)-whereas the object, when it exists, is an individual extramental thing-or more precisely, a series of individual things (e.g., all horses)-and thus is fully determinate (since Brentano is not a realist about universals, as is clear from Brentano 2011, p. 34 and1930, p. 74). I must confess that it is not clear to me why Brentano adds a question mark after the sentence "The content often indeterminate"; one explanation is that he is hesitant here about admitting determinate contents of presentation, and so is wondering whether the content is indeterminate only "often" or rather "always". But since the idea has already appeared earlier in the text, and without a question mark, I am inclined to say that he does indeed admit determinate contents. 13 Importantly, this text mentions names. Now, in the Brentanian tradition, the meanings of names are necessarily contents of conceptual presentations, never sensory contents; this is explicitly defended at any rate by Brentano's faithful pupil Anton Marty, who was developing a philosophy of language inspired by his master (Marty 1940, pp. 116-117). Brentano himself distinguishes meanings of individual names (e.g., 'Socrates'), and meanings of common names (e.g., 'horse'); he holds that to proper names there corresponds a "determinate" or "individual" presentation, while to common names there corresponds an "indeterminate" or "general", that is, universal, presentation (Brentano 2011, p. 100). While this indeed indicates that the meanings of common names are conceptual contents-since they are universal presentations-Marty confirms that the meanings of proper names are also conceptual: they are definite descriptions, which are combinations of conceptual contents (see Marty 1908, pp. 438-439, n. 2). It seems to be these distinctions that the text quoted above is referring to, since it speaks of "indeterminate" contents of presentations, thus in contrast to "determinate" contents, and as noted above, it mentions names; so the text seems to be about names and the determinate or indeterminate conceptual contents which constitute their meanings. In other words, even if Brentano does not say it explicitly, it is very likely that this text is only about conceptual presentations, whether individual or universal. This would fit quite well with the appearance in Brentano's text of the medieval distinction between first and second intentions, which applies precisely to concepts. 14 Indeed, as noted above, the usual view among scholars is that medieval philosophers take intentions to be concepts; more precisely, first intentions are concepts of things, while second intentions are concepts of concepts. When reading the text quoted right above, one might think that a first intention is the concept of the presented, which would thus be a concept of things-that is, it would subsume things-whereas a second intention is the concept of content, and would thus be a concept of concepts, since it would subsume other concepts, namely, those referred to as "presented", or first intentions. This reading seems to be confirmed by the fact that Brentano distinguishes between man understood as a thing, 'man' understood as a name, and man taken as a concept, of the last of which he indeed predicates a concept of a concept, namely, being a species. Let us call this the "conceptualist reading", which takes both kinds of intention, first and second, to be concepts for Brentano, following the standard medieval account of the distinction.
It seems to me, however, that Brentano's account differs from this standard interpretation. Instead of this conceptualist reading, I am rather tempted by what I will call the "referentialist" reading. My hypothesis is that Brentano is not talking about concepts, but rather about the references of the names 'presented' and 'content', that is, the things that these names name; thus, what is presented (not its concept) is a first intention, while the content (not its concept) is a second intention. In other words, what Brentano is speaking of is not the concepts expressed by the names 'presented' and 'content', but rather what these names name, or refer to (hence the "referentialist" reading), that is, the things presented and the contents of presentation respectively. One argument in favour of this reading is that in the list quoted above Brentano is not distinguishing between kinds of concepts, but rather between various kinds of things: namely, names, presentations (i.e., mental acts), contents (of conceptual presentations, admittedly), and objects. Thus, I take him to mean here simply that what is presented is a first intention, while the content is a second intention. Another clue in favour of my referentialist reading is this: in Brentano's framework, saying that the concept content is a second intention in the standard medieval sense would mean that the concept content is a concept of concepts; but since concepts for Brentano are mental contents (contents of presentations), this claim would not be very informative, since it would mean only that the concept content is about contents, just as if someone were to say that the concept concept is about concepts. It would simply be saying, for example, that the concept content applies to contents such as (the contents) man or socrates. (Compare this to the claim that being a species is a concept of concepts: this is much more informative, for it tells us that being a species does not apply to extramental things.) While such a point might be relevant to certain other discussions, it is hard to see what it would be doing in Brentano's text. I therefore tend to think that the standard scholastic, conceptualist understanding of the distinction between first and second intentions is not applicable to Brentano's text, and I will try to develop my referentialist reading.
Let me now quote-still in the same text as above-the second occurrence of the scholastic distinction, which in fact refers only to second intentions, and not to first intentions. The text aims to further describe those elements among the representations mentioned in the previous list that are exclusively related to mental acts of presentation (probably only conceptual presentations) (Brentano,Ps 76,n.  Surprisingly, while "the presented" was described in the first list as a first intention, it is here described as a second intention. Is there a sheer contradiction? I don't think so. What Brentano means is that, usually, "the presented" refers to the object of presentation, that is, one or several (possibly existing) external thing(s), for example, Socrates or horses; but the label can also be used to refer to the content itself of the act of presentation, that is, Socrates-taken-as-a-mentalimage or horse-taken-as-a-mental-image (an "image of presentation", as the text says). In other words, among the various labels that one can give to mental content, 'presented' would work, but only if one adds 'as a second intention', that is, precisely when treated as a content.
There is a third occurrence of the distinction, in a brief text which follows the two already quoted (Brentano,Ps 76,n.  Doesn't the text demand a conceptualist reading? Isn't it saying something like: "Distinctions among concepts of things divide things and point out their relations, while distinctions among concepts of concepts divide contents and point out their relations"? I think the referentialist interpretation would be perfectly fine as well, reading the text as follows: when you are distinguishing among first intentions, you are distinguishing among things themselves and you are picking out their relations, whereas when you are distinguishing among second intentions, you are distinguishing among mental contents; for example, a distinction among first intentions is one in which you are distinguishing man and horse, whereas a distinction among second intentions is one in which you are distinguishing determinate and indeterminate contents.

The Distinction and Brentano's Overall Theory of Intentionality
Now that I have presented my general understanding of Brentano's borrowing of the scholastic distinction between first and second intentions, let me see more precisely how this distinction works in his account of intentionality. This requires taking a closer look at the various items he describes in his lists. What exactly is a first intention, that is, "the presented" (das Vorgestellte)? We have seen that Brentano (Ps 76, n. 58723) equates this with the "object of presentation" (Vorstellungsgegenstand), and also calls it "the named" (das Genannte). As I said above, my view is that this is one or many extramental individual object(s) which (possibly) correspond to the presentation, for example, Socrates or horses. The fact that the presented is an extramental thing is confirmed by the equivalence between "presented" and "named". What Brentano calls "the named" is made clear in his logic lectures EL 80, where it turns out to be an extramental object, as Brentano (2011, p. 35) explicitly claims: The name designates in a way the content of a presentation as such, the immanent object. In a way, that which is presented through the content of a presentation. The first is the meaning of the name. The second is that which the name names.
[…] It is that which, when it exists, is the external object of presentation. One names by means of the meaning.
What we also clearly see in this text is that the "meaning" (Bedeutung) of a name, as Brentano says, is strictly to be distinguished from the named, a distinction which seems to be equivalent to that found in the Fregean tradition between sense and reference (even if senses in Frege, in contrast to Brentano, are not in the mind). For common names, obviously, there are multiple "named" extramental individual objects (e.g., all horses for the name 'horse'), while there is just one for proper names (e.g., Socrates for 'Socrates').
This text also helps us to better see what a second intention is, that is, the content of presentation. It is a mental entity, which is given as soon as the act is given. It is a sort of mental item "through" which, as Brentano says, the external object is presented. Importantly, the "content" (Inhalt) is also called "immanent object" in this text, just as in the Psychologie (Brentano 1924, pp. 124-125), where "content" (content in general, not just conceptual content) and "immanent objectuality" are identified. It is highly probable that Brentano maintained the equivalence between content and immanent object at the time when he wrote about first and second intentions, that is, in 1875. This can be reconstructed from the chronology of his theory of immanent objects. While the passage from the logic lectures where content and immanent object are identified is hard to date precisely-as stated in Rollinger's editorial introduction, the lectures were first given in 1869/1870, but regularly revised thereafter-Brentano was still speaking of "immanent objects" in his 1890/1891 lectures on descriptive psychology (see Brentano 1982, pp. 21-22). The usual date given for the rejection of immanent objects is 1904 (documented in Brentano 1930 and1952). One cannot rule out that Brentano was excluding immanent objects from his theory in 1875, given the absence of any explicit mention of them in the text in question; this would be surprising, however, given his acceptance of these objects in 1874, in 1890/1891, and up to 1904. Moreover, even if in 1875 the contents are not called "objects", they are very similar to (immanent) objects in Brentano's sense. Indeed, Brentano often stresses that he uses the term 'object' in a technical sense which is first found in the medieval tradition, according to which an ob-iectum is relative to a mental act and is that towards which the act is directed; the Latin etymology suggests that it is something "thrown" (-iectum) "in front of" (ob-) some other thing (see, e.g., Brentano 1952, p. 340). 16 Now, in 1875 the content is described as an "image", something which seems to be an "object" in Brentano's sense, that is, a representation to be contemplated by the mind; it is also called a "presented as presented", that is, something which, obviously, is contemplated (i.e., "presented") by the mind. A last argument: in the logic lectures, Brentano seems to use 'presented as presented' as a synonym of 'immanent object', since both expressions refer to the content of presentation (compare Brentano 2011, pp. 35 and 28). So in what follows, I will accept that in 1875 Brentano treated content and immanent object as identical.
Be that as it may, the external object can be absent, as is clear from the 1875 manuscript. 17 The external object thus stands in contrast to the act and its content, which "always [exist]". The point is also confirmed in a passage already quoted above from the same manuscript (Brentano,Ps 76,n. 58723): Each presentation has a content of presentation, but not every one has an object: objectless The idea is basically this: mental acts, such as a presentation of Socrates or a presentation of Lao Tzu, must have a content which explains their specific directedness, and this is independent of the existence of Socrates or Lao Tzu, since the acts will still have a different directedness even if Socrates and Lao Tzu do not exist; however, it can be the case that Socrates and Lao Tzu do exist, in which case the acts will also have a "presented" or an "object of presentation", as Brentano calls it, which is to be distinguished from the content.
Thus, the equivalences seemingly are: (presumably conceptual) content = immanent object = second intention (= meaning) on the one hand; and presented = external object = first intention (= named) on the other hand. The overall picture that this model gives us is the following (see Fig. 1). (While the mental act and the content, or immanent object, always exist, I use a dotted line for the external object to indicate that it might be non-existent. Recall that for universal presentations, there are possibly many external objects, but I leave aside this variant in the figure.) But what is the connection between these items and intentionality? What exactly is the act directed at? At the immanent object, the external object, or both? As pointed out by Chrudzimski (2001), Brentano suggests in some texts that mental contents, or "immanent objects", are given to us only in reflexive intentionality, that is, what Brentano calls "inner perception". 18 The idea is that the content, or immanent object, is accessible only to reflexive intentionality, which is directed towards both first-order intentionality and its corresponding content, while first-order intentionality is directed towards external things, although this is done "through" the immanent object, as Brentano also says. The point would thus be that there is indeed cognitive access to mental contents, or immanent objects, but it is given only at the reflexive level; first-order intentionality would thus be able to grasp the external object. However, the fact that first-order intentionality somehow goes "through" the immanent object preserves the thesis-already found in Brentano's Psychologie-that intentionality is a relation to an immanent object; that is, this object must somehow lie "on the path" traced by the intentional relation. For as already noted, Brentano uses the term 'object' in a technical sense to mean something towards which a mental act is directed, so an object is something which must somehow be thought of. One might want to say that in Brentano's account, the intentional relation is indeed a relation to the immanent object, but this object is nonetheless "phenomenologically transparent" (see Chrudzimski 2001, p. 106, among  16 On the medieval notion, see Dewan (1981). 17 Note that this also has a medieval parallel: some scholastic authors, notably Hervaeus Natalis and Peter Auriol, hold that acts of cognition always have an (immanent) object, even in cases where nothing corresponds to it in reality. For a recent comparative discussion of these two authors, see Klein (2020). 18 See Brentano (1982, pp. 21-22), among others. others); that is, first-order intentionality does not grasp it, but goes "through" it. That which grasps the object is reflexive intentionality. For example, when I think of Socrates, I do so by means of the immanent object socrates; what first-order intentionality is about is Socrates himself, which is nonetheless intended "through" the immanent object socrates, while the immanent object itself is given to reflexive intentionality only. To sum up, this complex structure would make it possible to allow immanent objects in the philosophy of mind, while at the same time making room at the epistemological level for cognitive access to reality (see Fig. 2) 19 . (The part with the small dots on the arrow is there to point out that the arrow somehow goes "through" the immanent object; the dotted line at the right end of the arrow is there to indicate that the relation to the external object might not hold, given that this object might be non-existent.) This detailed exploration of the interaction in Brentano between immanent objects, or "contents", and external objects, or "presented things", allows us to better understand how the distinction between first and second intentions works in his theory of intentionality. The reason why the content is described as a second intention while the external object is described as a first intention is this: the content, or second intention, is given to reflexive intentionality only, that is, to second-order intentionality, which is directed towards another intentionality, namely, a first-order (presumably conceptual) intentionality and its correlative content; this first-order intentionality is in turn directed towards a presented thing, which is thus aptly described as a first intention. In short, Chrudzimski's reading of Brentano's theory of intentionality supports and clarifies my referentialist interpretation of Brentano's use of first and second intentions: first intentions are what is grasped by first-order (conceptual) intentionality, that is, they are what is primarily intended, while second intentions are the contents of firstorder (conceptual) intentionality, and these contents are what is grasped by second-order intentionality, that is, they are what is secondarily intended. 20 The distinction is not about two kinds of concept, but about two kinds of thing, namely, external objects and mental contents.
Note the interesting parallels: in his lectures on descriptive psychology, Brentano (1982, pp. 22-25) calls nonreflexive, externally directed intentionality a "primary psychical relation" (primäre psychiche Beziehung) and inner perception a "secondary psychical relation" (sekundäre psychiche Beziehung); in his Psychologie, Brentano (1924, p. 180) calls the object of outer perception, such as a colour or sound, 21 the "primary object" (primäres Objekt), and the object of inner perception, that is, the act itself of outer perception and its content, its "secondary object" (sekundäres Objekt). 22 All this echoes my idea that first intentions are what is primarily intended, while second intentions are what is secondarily intended. Note in addition that for  (Figs. 2, 3, 11, and 12) are closely based on figures already found in Taieb 2017. Note that in order not to overload the following figures, I will omit the small arrow which goes from the act to itself; it should be kept in mind, however, that the intentionality by which the immanent object is grasped is reflexive intentionality and thus is also at the same time directed towards the act itself. 20 Thus, first and second intentions are "intentions" not in the sense that they are intending, but in the sense that they are intended. 21 There is a difficult point which I cannot really address here, for it would require a long discussion in its own right. Brentano is not a realist about secondary qualities. He clearly holds in his Psychology that these entities do not exist in reality, but only intentionally (see, e.g., Brentano 1924, pp. 28 and 129). Does this mean that the primary objects of outer perception are immanent objects (at least in the Psychology)? I am inclined to answer negatively. Brentano can be found to say that secondary qualities present themselves to us as real (Brentano 1924, pp. 129-132). Thus, I still take him to mean that these things, when they are our primary objects, are given to us as if they were external objects. This warrants the thesis that first-order intentionality is directed towards external objects, though probably only from a phenomenological point of view. On these issues, I follow again Chrudzimski (2001, esp. pp. 104-107), who, however, focuses on later works of Brentano (notably Brentano 1982). 22 On the fact that reflexive intentionality is not just about mental acts, but also their contents, see Brentano (1924, pp. 179-180).
Brentano primary and secondary intentionality always come together, since every mental act is always also conscious of itself (Brentano 1924, p. 141-194). 23 Thus, the distinction between first and second intention, although rarely mentioned in the texts, 24 seems to be parallel to a central feature of all presentations, and thus seems to hold for all conceptual presentations, since they all have primary and secondary intentionality, and consequently something primarily intended and something secondarily intended. This also shows that, even if the distinction between first and second intentions probably applies only to conceptual presentations, it echoes a general stance found in Brentano, namely, the idea that while our first-order intentionality is about extramental things-which are our primary objectsmental contents are given to us in reflexive intentionality only-that is, they are secondary objects (together with the acts whose content they are). On this point, sensory and conceptual presentations go hand in hand.
In short, the medieval distinction between first and second intentions is used by Brentano to draw a clear contrast between our (conceptual) thoughts about objects in extramental reality and our thinking about (conceptual) mental contents themselves, and it echoes the more general Brentanian principle that mental contents are given to us in reflexive intentionality only.

A Critical Evaluation
Let me now evaluate first (and briefly) whether Brentano's account of first and second intentions is historically faithful to the medieval scholastics, and second (at greater length) whether it is philosophically plausible.

Historical Accuracy
As indicated above, the standard account of first and second intentions in medieval philosophy is that first intentions are concepts of things, e.g., horse, while second intentions are concepts of concepts, e.g., being predicable of a plurality. It was a commonplace in medieval philosophy since Avicenna 25 to hold that second intentions are the subject matter of logic (see Amerini 2011). In my "referentialist" interpretation of Brentano, however, I have claimed that first intentions are for him not concepts, but things, while only second intentions seem to be concepts, that is, contents of conceptual presentations. Moreover, in the interpretation that I have put forward, Brentano is not primarily interested in logic; rather, he uses the medieval distinction to explain our grasping of objects outside the mind as opposed to our grasping of mental contents. So at this point the reader will probably think that either Brentano is a bad medievalist, or I am a bad interpreter of Brentano.
However, it might be the case both that Brentano was a faithful medievalist and that I am understanding his view correctly. In fact, although there was a standard understanding of the distinction between first and second intentions, the topic was nonetheless hotly debated in the Middle Ages. Consequently, in parallel to the standard account of first and second intentions, there were alternative uses of this distinction among scholastic philosophers. For example, one finds in Robert Kilwardby 26 the idea that first intentions are "the extramental things themselves"-very much as in Brentano-while second intentions are "their modes of being", e.g., "universal", although it is not clear that these modes are inside the mind, in contrast to Brentano's second intentions. 27 Moreover, Aquinas, who, as Amerini states, does not use the terminology of first and second intentions in his theory of concepts, nonetheless distinguishes between "things primarily intellected" (prima intellecta) and "things secondarily intellected" (secunda intellecta). Now, "things primarily intellected" are extramental things, just as first intentions are in Kilwardby, whereas "things secondarily intellected" are grasped by reflexive intentionality and are "intentions following from our mode of understanding" (intentiones consequentes modum intelligendi). These intentions are concepts, presumably of extramental things. 28 At any rate, this shows that while "things primarily intellected" are extramental things, "things secondarily intellected" are mental contents. Note also that Aquinas states that, in contrast to extramental things, "intellected things as such" (res intellecta secundum quod huiusmodi), that is, extramental 25 See al-Shifāʾ, al-Ilāhiyyāt 1.2 (Avicenna 1960, p. 10). 26 See De ortu scientiarum 48 (Robert Kilwardby 1976, p. 157). 27 Here I follow and quote from Pini (2002, p. 29), who discusses Kilwardby's text. Importantly, in Kilwardby, just as in (my interpretation of) Brentano, first intentions are intentions to the extent that they are intended, not intending, since they are the extramental things themselves. 28 This is defended by Amerini (2011, p. 556b); contra, see Pini (2002, pp. 54-55), who holds that they are concepts of concepts. things insofar as they are in the intellect, are given to the intellect insofar as it reflects on itself. 29 These considerations recall Brentano, for whom the things primarily cognized are external things, while things secondarily cognized are those same things taken as (conceptual) mental contents ("the presented to the extent that it is in the one presenting", as Brentano puts it) and given to reflexive intentionality. When one reads the praise of Aquinas in Brentano (see, e.g., Brentano 1952, p. 291), one might think that he was following Aquinas's view on things primarily and secondarily intellected, but dressing it in a more standard medieval terminology. 30

Philosophical Value
Let me now come to the question of the philosophical value of Brentano's view on first and second intentions. Does his use of the distinction lead to a good theory of intentionality? In my opinion, an account of intentionality in which mental contents are immanent objects is a bad theory in itself; it also threatens to lead to a collapse of the distinction between first and second intentions in Brentano's sense. Now, as indicated, Brentano often treated content and immanent object as identical, and it is very likely that he held this view at the time he wrote the "Fragment from a psychology lecture" of 1875. I would prefer an account without immanent objects; however, I still find quite valuable Brentano's use of the distinction between first and second intentions, and the more general idea on which it is based, that the mental contents thanks to which we think about extramental things should be given to us at the level of reflexive intentionality only: that is, they are not primary objects, but secondary objects (together with the acts whose content they are). Accordingly, I am going to show why I think immanent objects should be rejected; then I will suggest an alternative view of intentionality; and finally I will explain how the distinction between first and second intentions (or more generally, between primary and secondary objects) would still apply to this alternative account. In the course of the discussion, I not only will criticize the view I attribute to Brentano, but will also examine some attempts to save his system (without claiming to exhaust all possible strategies), and will say why I think these attempts fail. In conformity with Brentano's views and terminology, in my evaluation of his theory I will speak of "first" and "second intentions", and will assume that this is primarily a discussion about conceptual presentations. Nevertheless, what I say in what follows can easily be extended to his account of sensory presentations. For, as I said above, the distinction between primary and secondary intentions echoes a general stance found in Brentano, namely, the idea that while our first-order intentionality is about things in extramental reality, which are our primary objects, mental contents are given to us in reflexive intentionality only, that is, they are our secondary objects.
To begin, recall Brentano's position. There are two objects, an immanent and an external one, and the immanent object is somehow "phenomenologically transparent" (Chrudzimski 2001) for first-order intentionality; hence, the grasping of external things is possible (see Fig. 3).
The problem with this view is that if the immanent object is transparent, it is not at all clear to what extent is it still an "object", which in Brentano's sense (Brentano 1952, p. 340) is something towards which a mental act is directed. After all, its transparency implies that the act does not grasp it, but literally goes "through" it (see Fig. 4).
One answer to the problem would be to say that the immanent object is still somehow grasped, or "intended", by first-order intentionality (see Fig. 5).  , e.g., In Sent. I, d. 2, q. 1, a. 3, sol. (Thomas Aquinas 1929, p. 67), discussed in Pini (2002. 30 Note that the young Brentano, when he was planning to write a dissertation on Suárez, was already interested in first and second intentions, and held that concepts, when they are taken as objectsthat is, when they are themselves thought about-are grasped reflexively or "under second intention" (sub intentione secunda). Here, while an intention presumably is something intending, not intended, it is interesting to see that Brentano is already saying that concepts are given to reflexive intentionality. See Baumgartner and Hedwig (2017, p. 169), quoting from Brentano's ms. "Über die verschiedenen Arten von Distinktionen" (Fr Schr 8). I thank Martin Klein for the reference.
However, if the immanent object is indeed grasped by first-order intentionality, it seems that cognitive access to reality is threatened. Moreover, Brentano's distinction between first and second intentions entirely collapses, since the same entity is intended by both first-order and second-order (or reflexive) intentionality. In addition, in this new scenario, there is a new character whose identity is unknown, namely, the relation of the immanent object to the external object. The exact nature of this second relation requires explanation. 31 Independently of this, however, what remains clear is that this relation is not intentionality, since the relation of intentionality has as its foundation a mental act, not an object. But then, in this variant of the theory, it is not intentionality which is related to external things (see Fig. 6).
Another alternative would be to adopt a sort of two-step process by holding that first-order intentionality both grasps the immanent object and then, on the basis of this original grasping, turns itself towards the external object (see Fig. 7).
But this leads to there being two first-order objects for one and the same mental act, which is clearly contrary to experience. Moreover, this would be a bizarre account, for it would posit a double intentionality at the first-order level, with two primary objects (see Fig. 8).
Let us continue with the rescue mission. Given the previous worries, one might perhaps still want to maintain that there is a two-step process, but hold that the mental act first grasps the immanent object by reflexive intentionality and then, thanks to this internal grasping, directs itself towards the external object (see Fig. 9).
However, this would lead to the odd result that the directedness of first-order intentionality is provided by reflexive intentionality. Now, no matter how exactly this would happen, reflexive intentionality is usually posited only to explain how we grasp the underlying first-order mental activity as it is, not to provide it with its directedness. Note also that in Brentano, the immanent object must somehow also be the object of first-order intentionality (as is clear from the  31 Brentano sometimes defends a view very much like the one represented by Fig. 5, and tries to explain the nature of the relation between the immanent and the external object as one of "quasi-identity"; however, this seems to me to be simply a label for the problem rather than a solution. See Brentano (EL 72, nn. 12542-12543) and, for an analysis, Taieb (2018, pp. 151-152). intentionality passage from Brentano 1924 mentioned above, among others), and this would no longer be the case in this last scenario. Finally, this would lead to a sort of terminological confusion, if not also a conceptual one, since the firstly grasped object would be the so-called "second intention", while the secondly grasped object would be the so-called "first intention" (see Fig. 10).
What then should we do? I think the thesis that there is something like mental content which provides mental acts-whether conceptual or not-with their intentional directedness is a very plausible claim, which is widely shared among philosophers and can reasonably be maintained. 32 For example, when I think in turn of Socrates and of Lao Tzu, there must be some part or feature in my acts which make the first act be about Socrates and the second about Lao Tzu, and this part or feature is independent of the existence of Socrates and Lao Tzu, since I can think of them even if they do not exist (as is the case now). At the same time, the idea that mental contents are given to reflexive intentionality is also an interesting path to follow; or at any rate, as long as we hold that mental contents are directly given to us in our experience, 33 they should be considered the objects of reflexive intentionality rather than of firstorder intentionality, since otherwise our externally directed acts would turn out to be about internal mind-dependent entities, with the threat that we find ourselves "encapsulated" in our minds. To take again the same example, although there is something in my acts which makes one act be about Socrates and another about Lao Tzu, this something is not what my acts are about, for they are about Socrates or Lao Tzu; and this is again independent of the existence of Socrates and Lao Tzu. Thus, the result is this: mental contents are to be admitted, but should be removed from the path traced by first-order intentionality, that is, they should not be treated as its objects (i.e., as "thrown" [-iectum] "in front of" [ob-] it). 34 The picture will thus be the following (see Fig. 11). The opponents of the view are mostly disjunctivists, who have had a significant impact on contemporary discussion, though their position is by no means dominant. In any case, their theory is about perception, which is just one sort of mental act. For more on disjunctivism, see Soteriou (2014). 33 An interesting position is defended by Meinong, who says that mental contents are inferential objects which are not given to experience (see Meinong 1971, pp. 384-385, quoted anddiscussed in Marek (2001, pp. 268-269 and271-272).
Not only is this theory better from a general philosophical point of view, but it is also less ambiguous with respect to Brentano's distinction between first intentions and second intentions, or more generally between primary objects and secondary objects. Indeed, the content, which in Brentano's framework is a second intention, or more generally, a secondary object-that is, something which is intended by second-order, reflexive intentionality-in no way lies on the path traced by first-order intentionality. It is really thought about only at the reflexive level. So here there is a sharp contrast between first and second intentions, or primary and secondary objects.
The consequence of this alternative is that it leads to a non-relational account of intentionality, in other words, an account in which intentionality is not a polyadic property, but a monadic one. Indeed, as long as you have intentional objects, you can say that thoughts about non-existent objects are still relational (i.e., polyadic), since they are relations to immanent objects (even if this jeopardizes the distinction between first and second intentions, and thus should not be accepted by Brentano without qualification) (see Fig. 12). However, once you renounce immanent objects, then your picture of intentionality, at least for thoughts about nonexistents, becomes monadic and looks like in Fig. 13.
One option for a non-relational account of intentionality, perhaps the most famous one in contemporary literature, is adverbialism (see Kriegel 2011). On this view, to think of cats for example, is to think cat-ly: it is a way of thinking, as dancing the tango is a way of dancing. In the grammatical construction 'dancing the tango', 'the tango' is the object of the verb, which would usually mean that it refers to something on which some other thing is acting; but in fact, the phrase is specifying the kind of dancing, by adding to it something like a mode. Similarly, thinking of cats is not a relation to an object, but is a specification of the kind of thinking, or a mode of thinking. Note the important distinction: cat-ly is not an immanent object, it is a way of being of the act; it could still be described as a part of the act, but it is in no sense intended by first-order intentionality. You can still speak of some sort of content in adverbialism, namely the cat-ly part in thinking cat-ly (which makes your act be about cats and not, e.g., about dogs), and it could still be described as a "second intention", or a secondary object, provided it is given reflexively. Some authors hold that in fact the late Brentano had an adverbialist theory of intentionality (e.g., Chrudzimski 2001 and Kriegel 2016), which would mean that he held a view like that in the last figure, that is, a non-relational view. Note, finally, that I have argued above that Brentano in the 1875 text holds that contents are immanent objects, which seems to entail a relational theory of intentionality, because every act is at least related to an immanent object (as in Fig. 12). However, if it should reveal that he was not equating contents and immanent objects, then he perhaps briefly defended a non-relational account of intentionality around 1875 as well, in which contents are not intended by first-intentionality, but somehow embedded in the act (as in Fig. 13); I say "briefly", because immanent objects have clearly returned in his 1890/1891 lectures on descriptive psychology (Brentano 1982, pp. 21-22, quoted above) and are definitively abandoned only in 1904 (see again Brentano 1930 and1952).

Conclusion
Brentano's theory of intentionality clearly relies on medieval scholastic philosophy, both terminologically and philosophically. This holds for his description of the directedness of mental acts as an "intentional relation" and for his claim that intentionality is a relation to immanent objects with "intentional inexistence". A less well-known aspect of Brentano's theory is his use of the medieval distinction between first and second intentions, which was described in various ways in the Middle Ages, but which Brentano seems to understand in a way which recalls Kilwardby and Aquinas. Kilwardby does indeed claim that first intentions are extramental things. Aquinas further distinguishes between "things primarily intellected", which are extramental things, and "things secondarily intellected", which are concepts, presumably of extramental things; he also claims that we think of "intellected things as such", that is, extramental things insofar as they are in the intellect, via reflexive thoughts. For Brentano, first intentions are what is thought about by our first-order (presumably conceptual) intentionality, while second intentions are the content of our (conceptual) mental acts and are given to us reflexively. Brentano apparently uses this distinction to explain how we can still have (conceptual) cognitive access to reality while accepting immanent objects in our philosophy of mind: these objects are second intentions-that is, entities given to our reflexive, second-order intentionality-while first intentions-that is, what is primarily intended by our (conceptual) cognitive acts-are the extramental things. This is an interesting move, for it draws a clear-cut distinction between our thinking about extramental reality and our thinking about the mental contents thanks to which we are able to think about extramental reality: our cognitive access to the world is indeed due to mental contents, but these are not what our first-order intentionality is about. This indeed reflects Brentano's general stance, which can also be seen in his description of sensory presentations, that while our first-order intentionality is about extramental things-that is, they are primary objects-mental contents are given to us in reflexive intentionality onlythat is, they are secondary objects (together with the acts whose content they are). However, as I have tried to show, Brentano's own framing of the distinction is problematic: for if one accepts immanent objects, the overall system collapses, because either immanent objects are not intended by first-order intentionality and so are not objects, or they are intended by first-order intentionality and so are not second intentions (at least not exclusively)-not to mention the fact that the latter option threatens our cognitive access to reality. A better account, and one much more compatible with Brentano's distinction between first and second intentions, is to admit mental contents without treating them as objects.
Acknowledgements A first draft of this paper was presented in 2019 in Würzburg at the conference "Double Intentionality: Historical and Contemporary Perspectives". I thank the participants for their remarks, especially Alessandro Salice, Daniele de Santis, and Michela Summa. I also thank the editors of this special issue, namely, Martin Klein, Philipp Schmidt, and Michela Summa, for their very helpful suggestions on earlier drafts of this paper. Finally, I am grateful to two anonymous referees for their constructive criticism.
Funding Open Access funding enabled and organized by Projekt DEAL.
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:// creat iveco mmons. org/ licen ses/ by/4. 0/.