# Creating Artifactual Kinds ![rw-book-cover](https://readwise-assets.s3.amazonaws.com/media/uploaded_book_covers/profile_1188016/9ykkwDowTccPH9MfbJe6qZcQ8DZdRc0aaa56Gublvhk-cove_R4pH6wn.png) ## Metadata - Author: [[Jesús Vega-Encabo and Diego Lawler]] - Full Title: Creating Artifactual Kinds - Category: #articles - Summary: insert summary - My notes: - Summary: This text discusses how theories on artifact kinds address the creation of genuinely new artifacts from intentional creative processes. It highlights the challenges faced by intentional theories in defining the emergence of new artifact kinds based on new concepts. The authors argue that theories about artifacts must consider four key aspects to effectively address the ontology of artifacts. - URL: https://readwise.io/reader/document_raw_content/169323856 - Source File: Creating Artifactual Kinds by Jesús Vega-Encabo and Diego Lawler .pdf ## LLM Chats ## NotebookLM ## LLM Audio ## Highlights > In most cases, new artifacts bear strong resemblances to existing ones, being variations of previous artifacts, modified in their material or functional structure to adapt to new artifactual niches. But often artifacts are *genuinely new*, namely, original produc- tions showing a very weak family resemblance with the population of existing arti- facts or even lacking precedents of any kind in that population. ([View Highlight](https://read.readwise.io/read/01hwwwfrrd0k716zsgwz52e25c)) > In these cases, the creation of a novel artifact involves the appearance of a new kind of artifact that may present a new solution to an old problem or offer a new solution to a new problem. In many other cases, new artifacts are only *relatively new*, i.e., they are artifacts belonging to existing kinds but modified to varying degrees (modified in their mate- rial or functional structure, redesigned, adapted to new artifactual niches, etc.). These artifacts show strong family resemblances to certain artifacts which already exist, and they often represent alternatives to them. ([View Highlight](https://read.readwise.io/read/01hwwwg5j2b4e0szwmb1qm22bm)) > emergence of a *new* artifact depends on the possession of a *new* concept of that artifact, a concept that guides the production of that sort of object. ([View Highlight](https://read.readwise.io/read/01hx9wc68rgn0bpzh49x16ktzx)) > If this is so, then the intentional theorist faces a further dilemma: either she has to excessively weaken the conditions for possessing a concept of an artifact or she needs to concede that the constitution of the newly created kind *cannot* depend completely and effectively on the concept, because, especially in cases of creation, the creator cannot possess the concept that will guide the production of the new artifact ([View Highlight](https://read.readwise.io/read/01hx9wcpc481zmdd98ftpmj9da)) > the *creation requirement*, which demands an account of the appearance of genuinely new artifacts resulting from intentional creative processes. ([View Highlight](https://read.readwise.io/read/01je89j7pxe5e8m2pyma2v78wr)) > [The artifactual world is impressive in its novelty. A neophilic compulsion seems to characterize our contemporary culture,](https://readwise.io/reader/document_raw_content/169323856/#bookmark23)2 in comparison to ancient times where artifactual lineages were more stable and display less variability. With varying frequency, *any* technical culture witnesses the emergence of new artifacts. But what makes the appearance of *new* artifacts possible? ([View Highlight](https://read.readwise.io/read/01hwwwnfk7pgqgrkwbjdn73s6p)) > We have identi- fied a *factum* that requires explanation: *new* artifacts repeatedly appear in our world. A complete and plausible philosophical theory of artifacts and artifact kinds cannot ignore this fact of human culture. Taking this as a background condition, it is easy to see how it imposes a certain explanatory demand on theories concerning the ontological condition of new artifacts. A philosophical theory about the nature and identity conditions of artifacts and artifact kinds must apply to artifacts that cannot be identified as belonging to an existing kind because their very appearance marks the beginning of a new artifactual kind. Or to put it differently, any theory of this type must be able to account for the identity of genuinely novel artifacts. We call this the *novelty condition*. ([View Highlight](https://read.readwise.io/read/01hwwwr75hvfb71fa8akg1kyza)) > (1) if the theorist adopts a functional theory and chooses to account for the ontology of artifacts in terms of proper functions and, in turn, to account for those functions by referring to a causal reproductive history, her explanation will exclude an account of the nature of newly created artifacts; > (2) however, if she adopts an intentional theory and makes intentional dependence into a constitutive aspect of being this or that artifact, she must argue that the ([View Highlight](https://read.readwise.io/read/01je89yqrsr73z0gknhks40h90)) > Creation Requirement (CR): Any theory about the nature of artifacts must offer an explanation of the ontological nature of genuinely new artifacts that are the result of our creation. ([View Highlight](https://read.readwise.io/read/01hwwwsww7ftk2vcqhndyw7rzp)) > we identify an object as a corkscrew by specifying what the object is for, that is, for removing corks. We can then establish: “this new configuration of matter is a corkscrew.” If there are other previ- ous instances of this sort of object, then the novelty of this new arrangement of matter rests on the fact that it is a modified example of an already existing kind. ([View Highlight](https://read.readwise.io/read/01hx9xcqank15zvx0nehm26tqv)) > The appearance of the Newcomen steam engine is radically new. The introduc- tion of a new operational principle (the use of steam in a cylinder and atmospheric pressure after bringing about a vacuum) expresses a genuine novelty. Not even the fact that the use of steam was a previously known principle lessens its novelty, since it is the final disposition of the machine that is original ([View Highlight](https://read.readwise.io/read/01hwwwwsdsk5rfw6wck2kjj44m)) > We can all think of paradigmatic examples of innovations, outcomes of intentional creation that lack recognized lin- eages, such as the light bulb, the electric motor, or the microchip. Following this line of reasoning, it is certainly plausible to think that some sorts of technical development can only be described in terms of the emergence of a new kind of artifact. ([View Highlight](https://read.readwise.io/read/01hwwwwzzg05r90h81wm5awhr9)) > A new artifact must be characterized according to the identity conditions that are proper to it qua such an artifact and which constitute the kind to which it belongs. What relationship must exist between different individuals so they can be consis- tently grouped under one kind? Two responses are predominant in the literature on artifacts: you can identify the kind either by generalizing the traits of an individual with a certain function to other individuals with the same function or through the generalization of these traits to objects created with the same *intention* [(Grandy](https://readwise.io/reader/document_raw_content/169323856/#bookmark14) 2007, p. 29). Functional theories assert that sameness of function is the relevant criterion for kind membership. Artifactual kinds are primarily functional kinds, and, thus, the creation of a new artifact entails the creation of a new functional kind. An adequate description of an artifactual object would consist in the identification of a proper functional kind to which it belongs. In this view, then, each new artifact is *new* in virtue of its (proper) function ([View Highlight](https://read.readwise.io/read/01hwwx1e1x3qhw8e80kaw91z0a)) > *CR* picks up the following fact: some genuinely new artifacts are the result of > [our creation. This fact has metaphysical relevance because the emergence of a genu- ine new artifact also means the introduction of a new kind.](https://readwise.io/reader/document_raw_content/169323856/#bookmark12)7 Since newly created artifacts cannot be placed within actual lineages or understood by reference to these lineages, it seems impossible to determine their identity and persistence conditions in these terms. That is, we cannot establish their ontological nature by reference to the identity conditions of other artifacts. Newly created artifacts establish identity conditions for objects that are added [to our ontology. Any adequate metaphysical theory about the ontological nature of artifacts must be able to account for the iden- tity conditions of genuinely novel artifacts – which, by being genuinely novel, do not belong to any existing artifact kind.](https://readwise.io/reader/document_raw_content/169323856/#bookmark23) ([View Highlight](https://read.readwise.io/read/01hx9xjryh8hn4en71jvz4dxgv)) > [The pluralistic approach of functions proposed by Beth Preston (](https://readwise.io/reader/document_raw_content/169323856/#bookmark16)1998[,](https://readwise.io/reader/document_raw_content/169323856/#bookmark21) 2003[) follows this general strategy. According to her, we can attribute two sorts of functions to artifacts: proper and system functions (or systemic functions). As we have said, the proper function of an artifact is fixed by appealing to the causal history that led to the artifact being there with the structure that it effectively has – a history in which the intentions of design- ers and users obviously play a role. The proper function of an artifact is the “what for” of that artifact. The acquisition of these functions is a process analogous to the process by which biological features obtain their functions: a reproductive process contingent on success. On the other hand, system functions refer to tasks that the artifacts are able to carry out in terms of their materiality or the arrangement of their components. These depend on the capacities or dispositions that an artifact has and which may become manifest under certain circumstances, for example, when the artifact is employed in an idiosyncratic manner. Artifacts are intentionally propa- gated due to the (perceived) success that proper functions manifest in their actual use. However, systemic functions are the axis (Preston](https://readwise.io/reader/document_raw_content/169323856/#bookmark16) 1998, p. 250) on which a proper function is acquired for the first time, substituting the existing proper func- tion or adding a new one to it. ([View Highlight](https://read.readwise.io/read/01hwwx9n3gpnc88df79hwa2jyj)) > [The proper functions of an artifact (what the artifact does) are those functions that have been fixed through a history of selection that in a certain way fits the biological model of selection upon success. Etiological theories of function ascription fix proper functions by appealing to the causal history that led to the appearance of an artifact with the structure it effectively has. According to Vermaas and Houkes (](https://readwise.io/reader/document_raw_content/169323856/#bookmark2)2003[), these theories can be classified along two axes: on the one hand, according to whether the etiology of the item is reproductive or not and, on the other hand, according to whether the etiology is intentionalist or not. Prima facie, reproductive etiological theories of an intentional character seem more appropriate for artifacts, as they are able to accommodate intentional factors linked to the existence and persistence of artifacts. In principle, it would be implausible to admit a theory of artifact functions in which the causal history that fixes the particular function did not involve – at least partially – intentions. Furthermore, efforts to apply a reproductive model based on biological organisms in the field of artifacts have always encountered insurmountable difficulties, espe- cially when trying to explain what is involved in the processes of copying (Vermaas and Houkes](https://readwise.io/reader/document_raw_content/169323856/#bookmark2) 2003[; Soavi](https://readwise.io/reader/document_raw_content/169323856/#bookmark26) 2009). ([View Highlight](https://read.readwise.io/read/01hx9xrrpf3ts2bpf10dhagga2)) > Reproductive etiological conceptions seek an explanation of the emer- gence of new artifacts by appealing to the notion of proper function. In general, there is artifactual novelty when an artifact presents a new proper function. The key problem here is to determine the relevant causal history of new artifacts that have no reproductive ancestors. According to this model, novelties can be expressed as variations that are often introduced by the intentional and creative intervention of the designers, and afterwards, various selection pressures come to determine the success or failure of the new trait or artifact. This selection process is what deter- mines the proper function and identity of the artifact. > [However, there is a central tension at the heart of this view. On the one hand, the theory claims that novelties should be understood as variations resulting from inten- tional and creative intervention. On the other hand, it argues that the proper function is what gives identity and persistence to the artifact. But, if novelty emerges as the result of intentions, proper function cannot constitute the condition of existence and persistence, and the theory needs to appeal to other conditions.](https://readwise.io/reader/document_raw_content/169323856/#bookmark15)[10](https://readwise.io/reader/document_raw_content/169323856/#bookmark15) ([View Highlight](https://read.readwise.io/read/01hx9xtkpehnpv7ahewand8k5v)) > This process of artifactual selection appeals to the notion of systemic functions against a background of proper functions (i.e., once systemic functions become selected and fixed, they become proper functions). This position tells a story of intentional selection in cultural con- texts in which *exaptations*[,](https://readwise.io/reader/document_raw_content/169323856/#bookmark27)11 products of intentionally selected systemic functions, acquire different degrees of fixity until the artifact is reproduced resulting in a new proper function ([View Highlight](https://read.readwise.io/read/01hwwxbtqpgzsxsy591q01xjws)) > In the context of our argument, the most important conclusion about this approach can be expressed as follows: *CR* does not apply to this model of change through *exaptations*. On the contrary, the thesis that novelty is the result of an intentional selection process that operates on systemic functions goes against some of the assumptions that motivate *CR*. We could say that, in fact, this requirement ([View Highlight](https://read.readwise.io/read/01hwwxdee9p8stdv2rnkne6d0z)) > and its demands are not real concerns for this theory ([View Highlight](https://read.readwise.io/read/01hwwxd8ebpk1bmsaez691e5ha)) > Moreover, it faces difficulties for understanding it as a constraint on any theory about the nature of artifacts. ([View Highlight](https://read.readwise.io/read/01hwwxd5eyqa020svtb05x4n9w)) > However, if the novelty of a proper function is explained by an appeal to intentions, what prevents the intentions of *users* [from determining the new func- tions? Nothing seems to prevent it. Preston (](https://readwise.io/reader/document_raw_content/169323856/#bookmark21)2003[) has argued that there is not any phenomenological grounding for distinguishing between these two types of inten- tions. Her argument shows on the basis of empirical examples that the designer’s intentions do not differ in relevant aspects from the user’s intentions.](https://readwise.io/reader/document_raw_content/169323856/#bookmark22)12 The corollary of this discussion is that if the intentions of the designers are sufficient to assign a function to new artifacts (and to determine kind membership), so are the intentions of users. Thus, if a new use could be sufficient to attribute (or confer) a new proper function, then the distinction between proper functions and accidental functions would lose its meaning or collapse. ([View Highlight](https://read.readwise.io/read/01hwwxe2sbfc4v600kth6kxfn7)) > From the perspective of this pluralist functional approach, we can say that an > artifact is genuinely new when it carries out new functions, i.e., performs functions that did not exist before. The direct answer to the creative requirement *CR* is that a relevant causal reproductive history cannot be specified for genuinely new artifacts, *but* [this does not mean you cannot appeal to a process of temporal drift through which systemic functions are exploited to replace existing proper functions or to generate or add proper functions (Preston](https://readwise.io/reader/document_raw_content/169323856/#bookmark16) 1998, p. 250). This approach does not intend to explain novelty or creativity in the field of artifacts; on the contrary, it postulates a process of change in which different phases emerge against a back- ground of certain continuities. As a consequence, there is no creative production of a genuinely new artifact but a transition from one artifactual state to another in which functional properties are added, lost, or replaced. ([View Highlight](https://read.readwise.io/read/01hx9xzv4gbeh7x2asvmmga993)) > [Preston (](https://readwise.io/reader/document_raw_content/169323856/#bookmark21)2003[), in contrast to Vermaas and Houkes (](https://readwise.io/reader/document_raw_content/169323856/#bookmark2)2003), prefers to retain the normative and permanent character of proper functions and to introduce, as we have seen above, a new type of function to characterize artifacts: nonnormative, temporal systemic functions that account for the *nature* of new artifacts. However, this strat- egy can be taken as abandoning the novelty condition, since there would be no genuinely new artifacts; rather an artifact’s novelty is explained as the appearance of a systemic function that becomes stabilized as a new proper (normative) function after a period of time. She seems to challenge the demand of the principle of novelty (and this would extend to our version of the creation requirement *CR*) because she believes that a principle like this depends on an individualist misconception accord- ing to which the determination of a proper function, as *intended function*, may be in the hands of individual agents. ([View Highlight](https://read.readwise.io/read/01je8c4szyq7tcmx9nyj1xqf5n)) > Preston claims that artifacts are cultural products subject to processes of reproduction and use in a social context. Thus, she concludes that the distinguishing feature between a proper function and an accidental one does not lie in the intentional aspects as such but in the individualistic intentional aspects. But, regardless of whether you insist (correctly, from our point of view) that proper functions are standardized functions that involve long social processes of reproduc- tion and use, where the designers are not the only type of agents that contribute to the production of the artifact, the essential point is that new artifacts can only be properly characterized in terms of systemic functions, even if they are accidental, and these artifacts *could* ([View Highlight](https://read.readwise.io/read/01hx9y5xpwng13fsjyhfy250a1)) > For anyone who accepts *CR*, the conclusion is that *CR* forces to adopt different theories for standard artifacts and new artifacts. Of course, there remains the alternative of challenging the relevance of *CR* [for any theory on the nature of arti- facts. In any case, this theorist has to deliver a plausible philosophical account that makes room for novelty and creation in artifactual realms.](https://readwise.io/reader/document_raw_content/169323856/#bookmark17)[13](https://readwise.io/reader/document_raw_content/169323856/#bookmark17) ([View Highlight](https://read.readwise.io/read/01hx9y63pq51p9b57n3f6xq0st)) > *CR* [demands an account of both genuine novelties and the inauguration of new kinds and postulates that the determination of a proper function serves to define the conditions of identity and kind membership.](https://readwise.io/reader/document_raw_content/169323856/#bookmark2) ([View Highlight](https://read.readwise.io/read/01hx9y6q09yjk6dedf7derbwpm)) > we have argued that any conception of the nature of artifacts and artifactual kinds must be able to meet the challenge of characterizing the ontological nature of newly created artifacts. It was also suggested that the identity of proper function does not seem to be a good criterion to characterize them ontologically. It becomes very difficult to attribute a proper function to new arti- facts, the first members of a new artifactual kind, because this attribution must depend on a causal and reproductive history. ([View Highlight](https://read.readwise.io/read/01hx9yb9bmxd5ean5j9ch7sza6)) > Two routes seem to be open at this point. One of them might be to apply a different notion of function to prototypes and new artifacts – e.g., the notion of systemic function. However, one may wonder why, if this notion is enough to pick out a new artifact, it is not enough to identify what artifacts are in general – i.e., objects ontologically characterized by the fact that they perform a certain systemic function. ([View Highlight](https://read.readwise.io/read/01hx9ybp4z75kh0vsppws7whx0)) > What confers unity to the members of an artifactual kind? The answer is: the intention under which they have been produced. And this would be so for the *first member* [of the kind. However, once one has accepted that the proper function of an artifact is its intended function we have taken a first step away from a generalized functional theory towards a clearly intentional one. It would suffice for giving up a purely functional theory as a general analysis of the nature of artifactual kinds to realize that some artifacts (such as works of art) are not fundamentally characterized in terms of their function (intended or not) and that members of certain functional kinds have not even been intended as having the function that identifies them. Something can be a pencil even if its maker wants that it never be used as such. As Thomasson remarks having these cases in mind, “sharing an intended function cannot be essential to the nature of an artifactual kind” (Thomasson](https://readwise.io/reader/document_raw_content/169323856/#bookmark0) 2003, p. 594). ([View Highlight](https://read.readwise.io/read/01hx9yd4eh40hmndbgejyp3n5c)) > [What do we mean when we say that the intention of the creator of an artifact should involve an idea of the artifact in question? It is said that the creator must have a concept or “some description” (Hilpinen](https://readwise.io/reader/document_raw_content/169323856/#bookmark19) 1993, p. 157) of the artifact she will produce. This concept contains the set of properties (whether functional or material or others) that characterize the artifact and that will be imprinted on matter. The core of this concept is given by a description that relates to the kind of thing that is going to be created. Thus, if a maker intends to create a corkscrew, the contents of her intention will not only collect, for example, the following properties: “being able to remove corks,” “being of a white color,” “being a material resistant to tampering,” ([View Highlight](https://read.readwise.io/read/01hx9ygjyh1nx85xfb54ggan38)) > Let us suppose that it is sufficient to stipulate which aspects determine the kind. In this case, the success or failure of the execution and, therefore, the effective pro- duction of the artifact do not affect how the maker’s intention and the concept of an *A* that is involved in the intention are determined. At the same time, the maker seems to know what it is to be an *A*. It is purely contingent that the maker manages (or not) to succeed in carrying out the intention. ([View Highlight](https://read.readwise.io/read/01hx9ysvn74ekmextz9t049rxy)) > At this point, there are two additional problems. On the one hand, it is odd to say that the kind is fully determined prior to there being a member of it. This would seem to entail that there could be kinds whose extension is empty and has always been empty. This remark is not about the possibility of having a concept with an empty extension. We all know that there are many; the question concerns the possibility of *sortal* concepts grasped with independence of the existence of any member of the kind. ([View Highlight](https://read.readwise.io/read/01hx9ytw5by1vq475nk95vh0e5)) > Biological kinds, if they exist, could provide a good analogy at this point. We can have sortal concepts that identify members of extinct biological kinds (let us say, species), in a way that enables us to say that these sortal concepts have empty extensions. But in this case, those who grasp the concept have an idea of what it is to be a member of the kind. The question with the concepts of nonexisting artifacts is that our grasp of them does not give us a sufficient criterion to determine which sort of object would fall under its extension. ([View Highlight](https://read.readwise.io/read/01hx9yv41bzff7gc6155dfjhq6)) > Did Leonardo perhaps have a concept of something like a “plane” when he drew his sketches of flying machines? Were his *ideas* a way of delineating an artifactual kind by stipulating features ([View Highlight](https://read.readwise.io/read/01hx9yxbs3d8cwm7t35nsnd76s)) > relevant to *A*[? Here, intuitions could be in conflict. We claim that there is a clear-cut distinction between mere ideas of possible artifacts and those sortal concepts of artifacts that are needed to delineate a kind. A mere idea of an artifactual possibility, even if it is well defined and could itself be, in a sense, true of the artifact](https://readwise.io/reader/document_raw_content/169323856/#bookmark20)17 [(i.e., it would lead to an efficient realization if all technical and skill conditions were in place), does not constitute by itself the grasping of the sortal concept of an artifact or, at least, is not sufficient for possessing a concept of an artifact that will effec- tively and appropriately guide its production.](https://readwise.io/reader/document_raw_content/169323856/#bookmark22) ([View Highlight](https://read.readwise.io/read/01hx9yxk7q72r6r8y6rngzgvbd)) > If we choose to account for the nature of artifacts in terms of proper functions, the resulting explanation excludes an account of the nature of newly created artifacts. If we claim that the emergence of a new artifact depends on the possession of a new concept of that artifact, then we face a dilemma: either we need to excessively weaken the conditions for possessing a concept of an artifact or we need to concede that the constitution of the newly created kind cannot depend completely on the concept. We have argued that the reason for this dilemma is fairly simple: in the case of the creation of artifacts, the creators cannot possess the concept that will guide the production of the artifacts. ([View Highlight](https://read.readwise.io/read/01hx9z7k395svr9e0nvzpxqea1)) > • Creative production sets out from the idea of what we want to achieve; it starts from certain requirements that must be met and the satisfaction of which cannot be attributable to chance. > • Creative production is not guided, nonetheless, by a concept that describes the traits that determine the character of what is being produced, since genuine cre- ation institutes, so to speak, a hitherto nonexistent rule (it does not fit standards or concepts previously held by intentional agents). > • Creative production is done under certain conditions of ignorance in relation to the created product. > • Creative production is not carried out under conditions of complete blindness, as intentional agents have skills, knowledge, and previous solutions (present in a concrete social and cultural setting) that may suggest ways to proceed and solve problems. ([View Highlight](https://read.readwise.io/read/01hx9z9f2kmezxv00dx5ftaxv3)) > The creator, in a sense, imposes the norm, but what the norm dictates is not intelligible without effective and real success. As we have seen, the intentional-historical account emphasizes the first aspect. The functional-reproductive conception emphasizes the second one. Like other philosophical challenges, striking a balance between both dimensions may demand a new way of understanding the ontology of artifacts. ([View Highlight](https://read.readwise.io/read/01hx9zbfrafjmdt4pcnjmfxkdh))