# Slides
## Slide 1
- **Title:** The Metaphysical Puzzle of Artifact Kinds
- **[[Key Question]]:** What grounds the existence and identity conditions of artifacts and artifact kinds?
- Artifact kinds (e.g. ships, computers) seem to lack clear identity criteria.
- Threatens their objective reality - are they "really real"?
## Slide 2
- **Title:** Contrast with Natural Kinds
- Natural kinds (e.g. tiger, mountain) seem to have objective identity criteria.
- Governed by [[natural laws]] specific to the kind.
- Laws specify generation, development, destruction of kind instances.
- "There are laws governing the persistence of monkeys, telling us how they are generated, develop, and ultimately die."
## Slide 3
- **Title:** Artifacts and Mind-Dependence
- Artifacts seem to be mind-dependent in a way natural kinds are not.
- Depend on intentional human "artificers and utilizers."
- But only causally mind-dependent, not constitutively.
- "An artefact could survive the demise of all intelligent life, but no artefact could come into existence without intelligent life existing."
- **Discussion Q:** How might causal mind-dependence still threaten the objectivity of artifacts?
## Slide 4
- **Title:** Against "Conceptualist Realism" About Kinds
- View that Ks exist iff we have a coherent, satisfied concept of Ks.
- This leads to a regress or global anti-realism:
- Satisfying a concept presupposes determinate identity facts.
- So it can't explain identity facts without circularity or regress.
- "But that presupposes that the things in question have existence and identity conditions too, which need to be explained and grounded in turn."
## Slide 5
- **Title:** Problems with Persistence Conditions for Artifacts
- Artifacts of one kind seem easily changeable into another.
- Example: A table morphing into a chair or stool.
- Does it cease to be a table, or become both at once?
- "Can a table 'morph' into a chair? It seems so – in fact, it can do so very easily."
- **Discussion Q:** To what extent does this problem generalize to all [[material objects]]?
## Slide 6
- **Title:** Failures of Sortal Essentialism for Artifacts
- Even denying artifacts have essential kinds doesn't help much.
- Living things retain identity through "sameness of individual life."
- But no clear, uniform cross-kind identity criterion for artifacts.
- "It is far from clear that there is any single comparable equivalence relation that can be appealed to in the case of all individual artefacts."
## Slide 7
- **Title:** A Special Case: Machines
- Some artifact kinds, machines, seem to avoid the identity problems.
- E.g. engines, clocks.
- Machines embody a "distinctive unifying principle of activity."
- This activity governed by objective "[[engineering laws]]."
- **Discussion Q:** What is [[the key difference]] between machines and other artifacts like tables?
## Slide 8
- **Title:** The Martian Thought Experiment
- Imagine a Martian discovering a machine after humans are gone.
- The Martian could recognize its inherent functional nature.
- "A Martian, visiting Earth after all human life had been extinguished, could surely recognize in a working piston engine an object that does something, by its own very nature."
- In contrast, a table or hammer would be a mere inert object to the Martian.
## Slide 9
- **Title:** Machines vs [[Mere Instruments]]
- Machines seem to be "subjects of" special [[engineering laws]].
- Tables, hammers are merely "subject to" general [[physical laws]].
- Machines have a real "nature" and principle of activity.
- Instruments are merely structured for our use - they "do nothing at all by their own very nature."
- **Discussion Q:** Is this a clear and defensible distinction?
## Slide 10
- **Title:** Breakdowns and Malfunctions
- Further evidence that machines have a genuine functional nature:
- We can sensibly discuss machines and their parts malfunctioning.
- I.e. failing to fulfill their proper role in [[the machine]]'s overall functioning.
- "Thus, just as organisms can suffer from diseases, so machines can suffer from breakdowns."
- Not so for a merely broken instrument.
## Slide 11
- **Title:** The Fuzzy Line Between Machines and Instruments
- Admittedly, there may be borderline cases.
- As with living vs. non-living things.
- But this doesn't undermine a real distinction in the clear cases.
- "We should not allow the existence of borderline cases to undermine our belief in real distinctions."
## Slide 12
- **Title:** Lowe's Moderate Realism About Artifact Kinds
- Some artifact kinds (machines) are plausibly real and objective.
- In virtue of having genuine, law-governed natures.
- But not all - many are just mind-dependent human classifications.
- "The metaphysics of artifacts is a subtle domain mixing realist and anti-realist elements."
- An important test case for general theories in metaphysics.
# [[Speaker Notes]]
## Slide 1
- **Lowe is probing the metaphysical status of artifacts and artifact kinds:**
- What makes it the case that an artifact exists, persists, and is the very artifact it is?
- In virtue of what are two artifacts the same or different?
- Many familiar artifact kinds (ships, computers, tables) seem to lack firm identity criteria:
- When does an arrangement of parts become or cease to be a ship or table? How much can one change while remaining the same ship or table?
- This calls into question whether artifacts enjoy a mind-independent, objective existence:
- Perhaps they are just subjective human classifications, not objective features of reality?
- Quine's famous dictum: "No entity without identity" - indeterminate identity threatens real existence.
- Investigating artifacts is important for applying and testing general theories in metaphysics:
- Theories of objects, persistence, essence, kinds, mind-dependence, etc.
- Artifacts are ubiquitous in human environments, so their status has major practical implications.
## Slide 2
- **Natural kinds like biological species and geological formations seem to differ from artifacts:**
- Their instances seem to have objective, mind-independent identity criteria.
- Plausibly grounded in the [[natural laws]] governing the generation, development, and destruction of kind members.
- E.g., there are specific biological laws defining monkey life cycles and identity conditions:
- When a new monkey comes into existence, how it develops [[over time]], when it ceases to exist.
- These aren't just arbitrary human classifications - they track real, independent features of monkey nature.
- So the identity of natural kind instances isn't just "up to us" in the way artifact identity might be:
- We can't decide by fiat or convention when a new tiger comes into existence or ceases to exist.
- This is determined by objective, empirical, species-specific biological laws and processes.
- The contrast with the seeming flexibility and mind-dependence of artifact identity is striking:
- There don't seem to be artifact-kind-specific laws dictating table or ship identity conditions.
- Their identity boundaries seem much more dependent on shifting human concepts and practices.
- So natural kinds provide an illuminating contrast case in considering the metaphysics of artifact kinds.
## Slide 3
- **A key apparent difference between artifacts and natural kinds is mind-dependence:**
- Artifact kinds seem to constitutively depend on human concepts, intentions, and practices.
- Their existence and identity conditions make essential reference to human "artificers and utilizers."
- After all, ships and computers are human inventions, not independent, natural parts of the world:
- We created them to serve our purposes, and defined what counts as a ship or computer.
- Their identity seems intimately tied to our shifting conceptual schemes and conventions.
- However, Lowe argues that on closer examination, the mind-dependence of artifacts is more nuanced:
- They are clearly causally mind-dependent, but not necessarily constitutively so.
- I.e., they counterfactually depend on minded beings to bring them into existence.
- But once in existence, they might be able to persist in the absence of any minds.
- Consider a world where all intelligent life is wiped out, but the artifacts remain:
- The human-created artifacts would still seem to be there, even without us around.
- The pyramids, if they outlasted us, would still be pyramids - that doesn't require our active conception.
- So perhaps artifacts have a kind of hybrid status:
- Their existence counterfactually depends on minded artificers, but not their continued identity.
- They are created by, but not constituted by, human [[intentional states]] and practices.
- This opens up some space for a moderate realism about artifacts, navigating between two extremes:
- A strong anti-realism on which they're purely mind-dependent fictions.
- A simple realism ignoring their close connection to human interests and conventions.
## Slide 4
- **Lowe critiques a view he calls "conceptualist realism" about kinds:**
- On [[this view]], to be a real kind just is to be the referent of a coherent, satisfied concept we possess.
- E.g., 'computer' refers to a real kind iff we have a coherent concept of computers that some actual entities fall under.
- The view is realist in holding that our concepts can genuinely carve nature at its joints:
- When our concepts are satisfied, they successfully track real, mind-independent kinds.
- But it makes the existence of kinds constitutively dependent on our conceptual schemes:
- Kinds are only real insofar as they answer to our current concepts.
- Whereas a stronger realism would allow for real kinds that outrun our concepts or that we fail to recognize.
- Lowe's key objection is that conceptualist realism leads to a problematic regress or anti-realism:
- For our concepts to be "satisfied by the world", there must be determinate facts about the identity of entities.
- I.e., there must already be a fact of the matter about whether a given entity falls under a concept.
- But then these prior identity facts can't themselves be grounded in our concepts without circularity.
- So conceptualist realism can't provide a complete, non-circular metaphysical explanation of identity.
- The regress worry is that it keeps deferring [[the question]] of identity conditions:
- If we try to explain the identity of Xs in terms of the identity of Ys, and Ys in terms of Zs, etc.
- Without an end to the chain, no real explanatory ground of identity facts is ever reached.
- The other horn is an anti-realist one:
- To end the regress, one might hold that identity facts "bottom out" in a conceptual scheme.
- So there are no scheme-independent facts about what entities exist or their identity conditions.
- This is, in effect, a kind of anti-realist "conceptual relativity" about ontology.
- So Lowe contends that conceptualist realism is an unstable position:
- It keeps deferring the question of identity conditions, never reaching an explanatory ground.
- Or it collapses into an anti-realism that gives up on mind-independent ontological facts.
- A modest mind-dependence may be acceptable, but one that still allows for objective, conceptually-invariant identity facts.
## Slide 5
- **Lowe highlights how radically flexible the persistence conditions of artifacts can seem:**
- Compared to the more stable persistence conditions of natural kinds governed by kind-specific laws.
- Consider how easily a table could be transformed into a chair or stool:
- Just by changing how the very same parts are arranged and used.
- There's no clear, objective, table-specific law dictating when an arrangement of parts is or ceases to be a table.
- And the change seems to be largely a matter of shifting human purposes and conventions:
- In effect, we can decide by fiat to turn a table into a chair or stool just by resolving to use it that way.
- Whereas no perceptual or intentional change in us could turn a tiger into a different animal.
- At the extreme, it can even seem indeterminate or ambiguous whether the table persists or is replaced:
- Has the original table merely morphed into a chair, or has it ceased to exist and been replaced by a new chair?
- Whereas there's arguably always a determinate fact about whether a given tiger continues to exist.
- These cases suggest that artifact identity is much more dependent on vague, shifting human norms:
- Rather than being an objective, mind-independent matter governed by stable artifact laws.
- This threatens the idea that artifact kinds mark out genuine, human-independent ontological joints.
## Slide 6
- **Even giving up on artifacts having essential kinds doesn't fully resolve the identity problems:**
- One might think artifacts can fluidly change their kinds while still determinately persisting.
- Then the problems would only undermine artifact kinds, not the reality of individual artifacts.
- But Lowe argues this is much more plausible for living things than artifacts:
- There's a single, clear cross-species identity condition for organisms: sameness of biological life.
- If an organism of one species could in principle metamorphosize into another species,
- There would still arguably be a determinate fact about whether the same individual life persists.
- But in the case of artifacts, we don't seem to have any clear identity principle that holds across kinds:
- No obvious analogue of "same ongoing life" to ground sameness of artifacts across radical transformations.
- Is there any clear fact about whether a specific table later becomes a particular chair?
- Threatens a robust realism even about individual artifact persistence, not just artifact kinds.
- So giving up on sortal essentialism for artifacts (i.e., denying they have essential kinds):
- Doesn't give artifact realists the same resources as it does in the case of organisms.
- Since artifacts don't seem to have a clear "bare particular" identity independently of their kinds.
## Slide 7
- **Despite the problems so far, Lowe thinks some artifact kinds, machines, are plausibly real:**
- Machines like engines and clocks have a special metaphysical status among artifacts.
- The key idea is that machines embody an objective "unifying principle of activity":
- I.e., an inherent, stable tendency to behave in certain ways and perform certain functions.
- This principle governs and explains the coordinated workings of the machine's parts:
- And crucially, this principle is not just a human projection, but is grounded in objective laws.
- Distinctively "engineering laws" that specify how the machine and its components behave:
- These are a subset of natural laws, just ones specifically concerning functional artifacts.
- So what unifies and identifies a machine is this law-governed principle, not just human intentions:
- The principle is certainly the result of human design, but not merely "in the eye of the beholder."
- It's an objective, mind-independent feature of a machine once it's up and running.
- This contrasts with other artifacts like tables that seem to lack any comparable inherent principle:
- The unity and identity of a table just seems to be a matter of our deciding to assign it a certain function.
- But the table itself has no deeper, stable, law-governed tendency to behave in any particular way.
## Slide 8
- **Lowe invites us to imagine a Martian anthropologist studying human artifacts after our extinction:**
- The idea is to separate artifacts' objective natures from our projections and classifications.
- In the case of a machine like an engine, the Martian could recognize its inherent functional nature:
- The Martian would see that the parts operate in a coordinated way to perform a clear function.
- And that this functioning obeys certain regular, non-accidental principles (engineering laws).
- So the Martian could grasp the machine's essential nature without knowing what we used it for.
- But a simple tool like a hammer would present a very different picture:
- The Martian would just see an inert hunk of wood and metal, with no evident inherent function.
- The hammer's status as a hammer depends entirely on our using it that way, not any inner principle.
- So the Martian could not recognize the artifact's kind or identity independently of grasping our intentions.
- The thought experiment aims to show that machines, but not instruments, have objective functional essences:
- These essences are grounded in inherent engineering principles, not just external human categorization.
- And this suggests that at least some artifacts (machines) are genuinely real, objective kinds.
## Slide 9
- **The key contrast is between being "subject to" laws and being "subjects of" laws:**
- Mere instruments like tables and hammers are just passively subject to general physical laws.
- How a hammer falls when dropped is explained by gravity, mass, air resistance, etc.
- But there are no hammer-specific laws governing its behavior qua hammer.
- Machines, by contrast, seem to be subjects of special engineering laws:
- Laws that specify how that kind of machine characteristically functions.
- The laws governing an engine are not just the general laws governing its physical components.
- But laws proprietary to that type of engine and its special functional organization.
- Another way to put it: instruments don't seem to have real "natures", machines do:
- Instruments only have a principle of unity and characteristic functioning relative to our classifications.
- But machines have inherent, law-governed principles of activity independently of what we call them.
- In this sense, machines have intrinsic functional essences, instruments only have extrinsic/relational ones.
- We might say machines are "artifacts with natures" whereas instruments are "mere artifacts":
- Both are mind-dependent in originating from human design and intention.
- But machines then take on a certain objective, mind-independent functional character.
- Whereas instruments remain metaphysically dependent on us throughout.
- So Lowe is proposing a major division within artifacts, not just a difference of degree:
- Some (machines) have genuinely real kind-essences, others (instruments) are just human projections.
- This supports a form of moderate realism, on which some but not all artifacts are truly real.
## Slide 10
- **Lowe offers another consideration favoring the machine/instrument distinction:**
- We can meaningfully speak of machines and their components malfunctioning, not just breaking.
- Malfunctioning involves failing to perform an intended function or role within a larger system.
- It's more than just being physically broken - it's being functionally defective.
- Consider a carburetor that is physically intact but isn't mixing fuel and air properly:
- It's malfunctioning in the sense of not fulfilling its proper functional role in the engine.
- This functional role is specified by objective engineering principles, not just human whims.
- But it seems strained at best to say that a hammer's handle is "malfunctioning":
- If the handle is broken off, we'd say the hammer is broken, not malfunctioning.
- Since a hammer has no deep, objective functional nature, only a superficial human-assigned one.
- The idea of malfunctioning seems to presuppose an inherent proper functioning:
- And this in turn points to an intrinsic, objective functional nature.
- The nature is still the product of human design, but is not constituted by human recognitions.
- It's a mind-independent functional reality, as shown by the possibility of unrecognized malfunctioning.
- Lowe sees this as analogous to the way organisms can objectively malfunction (be diseased):
- Biological malfunctions point to an underlying objective, mind-independent biological nature.
- Similarly, mechanical malfunctions point to an underlying real, mind-independent mechanical nature.
- So the malfunction argument aims to show that machines have deeper objective natures than instruments:
- These natures ground real kind-identities and support realism about at least some artifact kinds.
## Slide 11
- **Lowe acknowledges a potential weakness in his proposed machine/instrument distinction:**
- There are likely to be unclear, in-between cases that don't neatly fit either category.
- Not every artifact kind will be determinately either a machine kind or an instrument kind.
- The boundary between the two categories is likely to be vague and contestable.
- But he thinks this is not necessarily a threat to a robust distinction between clear cases:
- Most important metaphysical distinctions involve some degree of vagueness.
- E.g., the living/non-living divide is not perfectly sharp, but still marks an objective difference.
- As long as there are paradigm cases on either side, borderline cases needn't undermine the distinction.
- The point is not that there's a clear-cut mechanical test for sorting artifacts into two neat boxes:
- But that the machine and instrument concepts point to very different metaphysical profiles.
- One involving greater objectivity, mind-independence, and intrinsicness than the other.
- Even if these come in degrees and involve unclear intermediate cases.
- Additionally, vagueness may be more epistemological than metaphysical:
- I.e., the indeterminacy may be in our concepts and knowledge rather than objective reality.
- There could still be a precise fact about whether something has a genuine functional nature.
- Even if it's not always evident to us where the line falls in hard cases.
- So the fact that knives, say, seem to fall somewhere in between pure instruments and machines:
- Doesn't necessarily undermine the idea that some artifacts have objective natures and some don't.
- As long as the great differences between clear cases like hammers and engines remain.
## Slide 12
- **In summary, Lowe is advocating a moderate, selective realism about artifact kinds:**
- Some artifact kinds, especially machines, seem to have real, objective essences and natures.
- They embody mind-independent functional principles that ground their kind-identities.
- So a realist view of these artifact kinds and their instances is justified.
- But not all artifact kinds share this status - many are just relational, mind-dependent categories:
- Their unity and persistence depends entirely on our shifting practices and intentions.
- So an anti-realist view of these kinds, like mere instruments, is more appropriate.
- On this view, artifacts are not metaphysically uniform, but a "mixed domain" of reals and non-reals:
- This diversity results from artifacts' complex relations to human minds and practices.
- Some achieve a genuine objectivity, but others remain entirely mind-dependent.
- Appreciating these nuances is crucial for a clear view of the metaphysics of artifacts.
- More broadly, artifacts provide a rich test case for applying general theories in metaphysics:
- Theories of objecthood, persistence, essence, realness, mind-dependence, etc.
- The complex middle ground artifacts occupy puts pressure on simplistic metaphysical frameworks.
- Challenges us to recognize intermediate degrees of objectivity and mind-independence.
- Artifacts are also of great practical and ethical significance in human life:
- So getting clear on their metaphysical status is not just an abstract intellectual exercise.
- But important for understanding the status of the artificial entities that shape our lives.
- Further questions: how far can the machine/instrument distinction be extended beyond artifacts?
- Are there other "functional object kinds" akin to machines outside the realm of human-made objects?
- How much can we generalize Lowe's "moderate realism" to other philosophically contested domains?