# Slides ## Slide 1: Title - [[Functions and the Aesthetics of Technical Artefacts]] - [[Wybo Houkes]] ## Slide 2: Introduction - Paper examines role of function analyses in explaining [[aesthetic appreciation]] of [[technical artefacts]] - Main conclusion: "despite first appearances, so-called '[[Functional Beauty]]' accounts cannot derive strength from analyses of [[artefact functions]]" - "on the contrary, these analyses constrain the possibilities for developing a suitable, function-based account" ## Slide 3: Introduction cont. - Complications "directly related to three phenomena": - functional variability - artefact flexibility - malfunctioning - Approach: "conceptual-engineering" - do function concepts support [[aesthetic appreciation]]? ## Slide 4: Desiderata for Theories of [[Aesthetic Appreciation]] - Selectivity - Polarity - Correctness - Aesthetic Dependence ## Slide 5: Selectivity & Polarity - Selectivity: "accounts should not be so lenient that everything may be regarded as beautiful, or so strict that nothing may" - Selectivity "features in critical discussions, especially of accounts in [[everyday aesthetics]]" - Polarity: "an account should encompass negative as well as positive modes of appreciation" ## Slide 6: Correctness & Aesthetic Dependence - Correctness: "an account should provide grounds for considering some acts of appreciation more appropriate or correct than others" - Enables "reasonable discourse about contrasting judgements" - Aesthetic Dependence: requires accounts to "incorporate or, preferably, explicate the relation [[between aesthetic]] appreciation or qualities and non-aesthetic features" ## Slide 7: Insights on [[Artefact Functions]] - Desiderata - Four desiderata for [[function theories]]: - Proper-accidental - Malfunction - Support - Innovation ## Slide 8: Artefact Function Desiderata cont. - Proper-accidental: distinguish "stable or entrenched functions" vs "occasionally, transiently or even accidentally function" - Malfunction: allow that artifacts may "fail to realize" [[proper function]] - Support: "ground function ascriptions in non-functional features" - Innovation: "ascribe intuitively correct functions to innovative items" ## Slide 9: Archetypical [[Function Theories]] - Three main types: - Intentionalist - Causal-role - Selected-effects - "None of these archetypes meets all desiderata" - But "archetypes can be combined and developed to compensate each other's shortcomings" ## Slide 10: Sophisticated [[Function Theories]] - Examples of combined theories: - ICE theory (Houkes & Vermaas) - General-design theory (Krohs) - Pluralist account (Preston) - Theories diverge but "each theory strictly speaking meets all four desiderata for function theories" ## Slide 11: Tension in Function Theories - "it proves difficult to resolve fully a tension between two aspects of function ascriptions": - Stability - "rooted in relatively stable design and use practices" - Discontinuity - "breaks with tradition," innovation, improvisation, malfunction - Theories differ in emphasis but must accommodate both - "Incorporating all forms of flexibility, or even only all varieties of malfunction, is a balancing act" ## Slide 12: Functions Not Essential to Artefacts - "functions are found to be not as crucial to artefacts as is commonly but also cursorily assumed" - Functions don't pick out "reliable cluster of properties that could constitute the nature of artefact kinds" - Functional categorization of artefacts appears dominant - But experiments suggest "pluralism of categorization principles" may be at work ## Slide 13: Functional Complication 1 - Variability - Functional variability: artefact functions are multiply realizable, no reliable property clusters - Makes meeting FB account desiderata "Correctness" and "Support" difficult - E.g. winged and air-pump corkscrews equally functional - No basis for judging one more functionally beautiful ## Slide 14: Parsons & Carlson's Functional Beauty Account - Three modes of appreciating functional beauty: - Fit for form - Simplicity/elegance - Visual tension - But functional variability entails ## Slide 15: Complications for FB Accounts - Variability prevents perceiving simplicity/elegance or visual tension based on standard vs contra-standard features - Only "fitness for form" mode applies, based on variable features "somehow indicative of functionality" - But all items in broad category are equally functional - No basis for correctness of appreciation as more/less fit or beautiful ## Slide 16: Sauchelli's FB Account & Functional Variability - Sauchelli: function ascriptions trigger sets of expectations about how function is realized in form - Appreciation explained by match/mismatch to expectations - But experiments show expectations based more on familiarity than objective functionality - Undercuts aim of providing normative standards for appreciation ## Slide 17: Possible FB Responses to Variability - Potential moves: - Appeal to narrower functional categories with more constrained properties - But still unclear categories capture all key aesthetic features - Or identify "Goldilocks" categories between too broad and too narrow - Theories don't themselves provide these, or justify their unique aesthetic relevance ## Slide 18: Functional Complication 2 - Artefact Flexibility - Artefact flexibility: items' functions change over time, items support multiple functions - Makes correctness claims and selectivity difficult for FB accounts - Allows "evaluative retrofitting" - imagining some function to rationalize almost any aesthetic judgment - Proper functions provide some stability, but items may still be appreciated based on outdated or inapplicable functions ## Slide 19: Artefact Flexibility Examples - Plate armour appreciated for ceremonial vs protective function - Cassette players in a world without cassettes - Accounts need principled way to select between: - historically accurate but inapplicable functions - current functions - imagined functions - Selecting some without function-theory-based criterion makes functional beauty depend on external standard ## Slide 20: Functional Complication 3 - Malfunctioning - Function theories "have trouble fully accommodating malfunctioning" - Expect similar and additional challenges for FB accounts - Even malfunctioning tokens create issues - e.g. damaged Anna G corkscrew - Divergent judgments from focusing on different functional categories ## Slide 21: Malfunctioning Token Example - Damaged Anna G corkscrew still works with improvisation - Aesthetic judgments might: - Consider it malfunctioning qua Anna G, lacking standard features - Consider it now a T-shaped corkscrew, variable features interfering - Consider it still successful qua corkscrew - Each grounds different judgment and "aesthetic directive" - FB accounts must select one to meet Correctness and Selectivity ## Slide 22: Type Malfunctioning - "Anna G corkscrews, because of their structural capacities, perform their proper function poorly" - Two FB options: - Ignore lack of functionality, appreciate "formal aspects" - Violates Polarity desideratum - Factor in lack of utility - Requires a broader functional category for comparison - But broad categories lack standard and contra-standard features ## Slide 23: Phantom Functions - "Phantom functions" - items that "completely fail to realize their proper function" - Function theories allow ascribing never-realized functions - But phantom-functional items "do not wear their lack of functionality on their sleeve" - Seem incorrect to ground negative appreciation in unfit look - Denying phantom functions violates function theories and puts aesthetics before function ## Slide 24: Conclusions - Potential Paths Forward - Develop a function theory better capturing stability and flexibility - Consciously limit scope/desiderata of FB account - Treat functional beauty as one of multiple aesthetic modes - Overall: "functions are not as crucial to the aesthetic appreciation of artefacts as proponents of Functional Beauty accounts might like them to be" # Speakers' Notes ## Slide 2 - Houkes' paper takes a deep dive into the relationship between two fields: philosophical analyses of artefact functions and aesthetic appreciation of technical artefacts like corkscrews, furniture, clocks, etc. - His key conclusion, as we see here, is that despite initial appearances, function analyses don't actually strengthen accounts of how we aesthetically appreciate artefacts based on how well they serve their functional purpose. - In fact, he argues the complexities that come up in rigorously understanding artefact functions end up significantly constraining attempts to ground aesthetic judgments in functions. ## Slide 3 - Houkes identifies three key phenomena in how function theories handle artefacts that create complications for tying functions to aesthetics: - Functional variability refers to the fact that very different designs and structures can all serve the same function, so there's no single cluster of properties that reliably tracks function. - Artefact flexibility points to how artefacts can shift between different functions over time and how a single artefact can support many different uses. - Malfunctioning has to do with how artefacts can fail to serve their supposed function, either at the individual token level or at the type level. - To examine this topic, Houkes takes what he calls a "conceptual engineering" approach - looking at whether the concepts of function developed in philosophy are actually fit to do the work required by aesthetic theories. ## Slide 4 - Before getting into the details of function theories, Houkes lays out a set of four desiderata that he thinks theories of aesthetic appreciation should aspire to satisfy. - These spell out what it would mean for an aesthetic theory to be correctly attuned to our actual practices of aesthetic judgment. - He draws these criteria from various discussions in aesthetics, though not everyone endorses all of them. - We'll go through each of the four desiderata in a bit more detail on the next few slides. ## Slide 5 - The first desideratum is selectivity, which has to do with the scope of what a theory counts as legitimately subject to aesthetic appraisal. - To meet selectivity, a theory shouldn't be so permissive that it allows absolutely everything to count as beautiful or aesthetically valuable in some way. - But it also shouldn't be so restrictive that nothing can satisfy its standards. - Houkes notes this criterion is especially prominent in debates around everyday aesthetics, which aims to expand aesthetic discourse beyond just fine art. - The second desideratum, polarity, states that a good theory needs to make sense of both positive and negative aesthetic judgments. - It can't just focus on what makes things beautiful while having nothing to say about ugliness. - This is a mark against theories that try to define beauty in terms of some property but can't turn around and define ugliness in terms of the opposite property. ## Slide 6 - Correctness is the idea that aesthetic theories should support some aesthetic judgments being more appropriate or proper than others. - There should be standards that provide a basis for saying one aesthetic reaction fits the object better than another. - This allows for meaningful debate and disagreement about aesthetic reactions. - It's important to note that correctness is distinct from selectivity. - You could have a very permissive theory that still claimed proper standards, or a selective theory without real standards. - Aesthetic Dependence is the expectation that aesthetic properties depend in some way on non-aesthetic properties. - Theories should specify just how aesthetic qualities arise from things like an object's shape, color, texture, etc. - This is a distinct notion from Correctness, since you could have Correctness just in terms of relations between aesthetic properties. ## Slide 7 - Turning now to the philosophical work on artefact functions that Houkes is assessing. - He explains how trying to formulate rigorous theories of function has revealed challenging desiderata that theories aim to meet. - These form an important background for understanding the moves function theories make and why they create complications for aesthetics. - The first desideratum is the "Proper-accidental" distinction. ## Slide 8 - Proper-accidental refers to how function ascriptions indicate both a relatively stable, established, "proper" function, but also allow more transient, happenstance functions. - Like how a screwdriver's proper function is to drive screws but it can also serve in a pinch to pry open a paint can. - The second desideratum, Malfunction, points to how we need to make sense of artefacts failing to fulfill their proper functions. - The classic example here is how a broken car or clock is still seen as "for" transportation or telling time even though it doesn't actually do that. - The Support desideratum says that function ascriptions need to be grounded in something other than just more functions to avoid circularity. - There should be some non-functional properties that justify attributing a particular function and distinguish real functions from merely imagined ones. - Finally, Innovation requires that theories can assign intuitively correct functions to novel artefacts, not just defer to established categories. - For example, when aspirin was discovered to prevent blood clots, theories should allow saying it genuinely gained a new proper function. ## Slide 9 - With that background, Houkes identifies three main families or "archetypes" of function theories in the literature: - Intentionalist theories define the proper function of an artefact in terms of the intentions of designers, producers, and users - what they mean for it to do. - Causal role theories instead look to the concrete causal capacities of the artefact and the role it actually plays in some larger system. - Selected effects theories tie functions to the effects that led to the reproduction or proliferation of some type of artefact via some selection process. - Importantly, none of these theory types on its own satisfies all the desiderata we just saw. - They each face challenges with some subset of the criteria. - However, the different theories can potentially be combined to cover each others' weaknesses. ## Slide 10 - This leads to a few influential examples of more sophisticated function theories that try to integrate the archetypes we just saw: - The ICE theory from Houkes & Vermaas defines functions in terms of use plans from designers and users. - It satisfies the desiderata by allowing both proper and accidental functions, malfunctions, connections to physical capacities, and shifting functions through new use plans. - The General-Design theory from Krohs looks at how functions arise from components being arranged according to a "type-fixing" design. - Preston's Pluralist account makes use of both selected-effects and causal role notions of function for different aspects. - While these combined theories differ in their details, Houkes suggests they all do a reasonably good job of jointly satisfying the four function theory desiderata. ## Slide 11 - However, this is not to say these more sophisticated theories are without problems. - In particular, there is a key tension that remains difficult to fully resolve between two conflicting aspects of actual function discourse: - On one hand, proper function ascriptions seem to reflect relatively stable and consistent practices of use and design over time. - This stability grounds proper functions in objective features of artefacts, even if users aren't always consciously aware of the full history. - On the other hand, we see clear cases of discontinuity and breaks from conventional functions. - This shows up in innovative designs that introduce new functions, creative user improvisation, and various malfunctions. - The different theories vary in how much they emphasize stability vs flexibility. - But they all have to allow both to at least some degree to capture the nuances of actual usage. - Striking this balance is tricky - leaning too far to either side risks missing important phenomena. ## Slide 12 - This brings us to a general conclusion about the relationship of functions to artefacts: - Contrary to a simplistic view one might initially hold, functions don't seem to be strictly essential to the nature of artefacts. - It's not the case that there is a single simple function that all artefacts of a given kind must necessarily serve. - Functions don't pick out a stable set of properties that all and only members of an artefact kind always have. - We might think artefact categories are defined primarily by function, like corkscrews being for removing corks. - But experimental work suggests a more pluralistic view, where function is one important factor in categorization but not the whole story. - Houkes emphasizes that this more limited role for function is fully compatible with the sophisticated function theories he has surveyed. ## Slide 13 - With that background on function theories in hand, we can now appreciate the complications Houkes identifies for grounding aesthetic judgments in functions: - The first major complication stems from the functional variability we discussed earlier. - The fact that a given function can be realized by many different physical structures means functions alone don't pick out consistent clusters of properties. - This makes it difficult for Functional Beauty accounts to cleanly derive aesthetic standards from functions. - For instance, winged corkscrews and air-pump corkscrews have very different appearances but are equally good at removing corks. - There's no clear basis for judging one as more functionally beautiful than the other just based on its function. ## Slide 14 - To dig into this issue further, let's look at some details of a prominent Functional Beauty theory from Parsons & Carlson: - They analyze the aesthetics of functional objects in terms of three modes of appreciation: - "Fit for form" refers to perceiving an object as well-suited for its function, with no extraneous features detracting from that. - "Simplicity/elegance" is perceiving an object as having mainly features that suit its function, without many irrelevant features. - "Visual tension" comes from seeing some features that seem ill-suited for the core function but in fact serve it in a non-obvious way. - The problem functional variability poses for this account is that it undercuts the notions of standard, variable and contra-standard features it relies on. - For broad functional categories like "corkscrew", there aren't established sets of features everything needs to have or avoid to serve that function. ## Slide 15 - This functional variability directly precludes basing aesthetic judgments on perceiving functional objects as simple/elegant or as exhibiting visual tension. - Those modes of appreciation only make sense if some features can be classified as standard or contra-standard for the functional category. - With wide variability in form, the only mode of appreciation left is "fit for form" based on variable features that somehow seem to suit the function. - But without a baseline of standard features, it's unclear how to judge some variable features as better suited to the function than others. - All the different corkscrews are equally functional at the end of the day. - So functional variability undermines both the Correctness and Selectivity desiderata for this sort of functional beauty theory. ## Slide 16 - Sauchelli proposes a somewhat different functional beauty account that analyzes aesthetic judgments in terms of matches and mismatches between perceptions and functional expectations: - The idea is that when we consider an artefact in light of its function, we generate a set of expectations about how that function will be physically realized. - Aesthetic pleasure then arises from the object conforming to those expectations, while displeasure comes from it violating them. - This view can somewhat accommodate functional variability by allowing expectations to be object-relative. - However, Houkes argues this account still founders on the Correctness desideratum: - Empirical research on artefact categorization suggests functional expectations are heavily driven by familiarity, not objective functional optimization. - So this view may explain some of our aesthetic reactions, but it can't justify them as appropriate or inappropriate relative to real functional standards. ## Slide 17 - Houkes considers a few ways functional beauty accounts might try to get around the functional variability complication: - One option is to focus aesthetic analysis on narrower functional categories that place more constraints on physical implementation. - We might look at winged corkscrews specifically rather than corkscrews in general. - The trouble is that even these narrower categories likely still admit of significant variability. - Additionally, it's not clear the function-relevant features at this level will necessarily contain everything relevant to aesthetic assessment. - Another option is to seek out a middle level of specificity between extremely narrow and broad functional categories. - But as Houkes emphasizes, the function theories themselves don't identify what these Goldilocks categories would be or provide a rationale for their unique aesthetic relevance. ## Slide 18 - The second major complication Houkes discusses is the flexibility of artefact functions: - Over time, artefacts can shift away from their original functions and users can creatively impose new functions. - Any given artefact also supports multiple different uses and functions simultaneously. - This spells trouble for Functional Beauty accounts meeting the Correctness and Selectivity desiderata: - If functions can vary across time and multiple functions always coexist, it becomes possible to rationalize almost any aesthetic judgment in terms of some actual or hypothetical function. - Proper functions might seem to provide some grounding, but artefacts can still support aesthetic appreciation relative to outdated or inapplicable proper functions. ## Slide 19 - To make this more concrete, think of elaborate decorative suits of armour from the 16th century: - We might aesthetically appreciate their elegance and visual complexity in light of a ceremonial function, even if they were terrible for actually fighting in compared to simpler armour. - Or consider Sony Walkman cassette players in an age after cassette tapes have disappeared: - You could still appreciate their sleek design as fitting for a portable music player function they can no longer really serve. - To avoid this "evaluative retrofitting", Functional Beauty accounts need a principled way to pick out the specific functions that should be the basis for aesthetic judgment: - Should it be historically accurate but currently inapplicable functions, current functions, or potential future or hypothetical functions? - Whichever way they go, the function theories alone don't provide a clear criterion for this choice: - That means grounding aesthetic judgment in functions requires some other standard besides just pure function. ## Slide 20 - The final complication Houkes explores is malfunctioning: - He notes how even the most sophisticated function theories have difficulties fully capturing the complexities of malfunction. - If making sense of malfunctions is challenging for function theories in general, we can expect it to create issues for Functional Beauty accounts too. - There are complications both with how to aesthetically assess individual malfunctioning artefacts and with artefact types that tend to malfunction. - Even individual cases generate conflicting aesthetic judgments depending on the functional framing. ## Slide 21 - To illustrate this, Houkes gives the example of an Anna G corkscrew that has had its arms broken off: - With a bit of improvisation, a user could still get it to open bottles, just not as smoothly as intended. - Aesthetically, we might judge this object as: - Malfunctioning qua Anna G corkscrew, since it is now missing distinctive standard features. - Functioning but inelegant qua improvised corkscrew, since the broken arms now just get in the way. - Successful qua corkscrew simpliciter, since it still gets the job done. - Each of these judgments makes sense relative to a different functional frame, and would motivate a different course of action: - Repair the arms to restore function, remove them to improve function, or just leave it as is. - To meet Correctness and Selectivity, a Functional Beauty account needs to specify which frame should take precedence, but the function theories themselves don't clearly make that call. ## Slide 22 - Malfunctions can also happen at the type level, with some artefacts tending to serve their intended functions quite poorly: - Imagine some line of corkscrews that, due to how they are designed, break very easily or require excessive force to use. - For these artefact types, Houkes sees a dilemma for Functional Beauty accounts: - One option is to just ignore the dysfunction and aesthetically assess the artefacts only based on their non-functional aspects: - Parsons & Carlson describe "bracketing" utilitarian functions to focus on formal qualities as one mode of aesthetic appreciation. - But this strategy violates the Polarity desideratum by providing no way to ground negative aesthetic assessments in poor functionality. - The other option is to factor in the artefact type's subpar functionality, judging it as uglier because of how poorly it fulfills its function: - The challenge here is that making these comparative judgments requires a quite broad functional referent: - It only makes sense to call this corkscrew ugly compared to corkscrews in general, not to other instances of its specific malfunctioning type. - But recall that broad functional categories like "corkscrew" lack robust sets of standard features to ground aesthetic assessment: - So either way, artefact-type malfunctions pose a serious challenge for meeting all the desiderata identified at the start. ## Slide 23 - As a final wrinkle, Houkes points out the phenomenon of "phantom functions" - artefacts that are ascribed functions they uniformly fail to achieve: - Think of "weight loss" headbands or bracelets that have no actual efficacy, or heavily hyped tech gizmos that never deliver their supposed benefits. - Many function theories allow these artefacts to still be properly categorized in terms of their intended or imagined function: - How should a Functional Beauty account handle such cases? - It seems wrong to positively aesthetically assess these artefacts as well-suited for their supposed function, since they don't actually serve it at all! - But phantom-functional artefacts often still have design features that suggest or imply functionality: - They don't necessarily look dysfunctional, which would be needed to ground negative aesthetic assessment on functional grounds. - Houkes argues that insisting phantom-functional artefacts can't actually have the functions ascribed to them avoids these issues: - But it requires overriding the considered judgments of function theorists to put aesthetic intuitions first: - This seems to invert the ideal explanatory order, with aesthetics now constraining functions rather than vice versa. ## Slide 24 - Wrapping up, while the complications explored here don't entirely rule out Functional Beauty accounts, they do point to some significant limitations and challenges: - Houkes sees a few potential ways forward: - Developing a new function theory that more robustly captures the nuances of functional stability and flexibility. - Explicitly limiting the ambitions and scope of Functional Beauty accounts to avoid the most problematic cases. - Treating functionality as just one component of artefact aesthetics rather than the exclusive or primary basis. - Ultimately, Houkes suggests, grappling with all the complexities in how functions actually work reveals they can't carry as much weight in aesthetics as Functional Beauty advocates might have hoped: - Functions play some role in aesthetic appreciation, but they aren't the simple skeleton key they might have initially seemed: - Fully understanding the aesthetic dimension of artefacts requires considering non-functional factors as well.