ArticlePDF Available

Abstract and Figures

Semantic features are components of concepts. In philosophy, there is a predominant focus on those features that are necessary (and jointly sufficient) for the application of a concept. Consequently, the method of cases has been the paradigm tool among philosophers, including experimental philosophers. However, whether a feature is salient is often far more important for cognitive processes like memory, categorization, recognition and even decision-making than whether it is necessary. The primary objective of this paper is to emphasize the significance of researching salient features of concepts. I thereby advocate the use of semantic feature production tasks, which not only enable researchers to determine whether a feature is salient, but also provide a complementary method for studying ordinary language use. I will discuss empirical data on three concepts, conspiracy theory, female/male professor, and life, to illustrate that semantic feature production tasks can help philosophers (a) identify those salient features that play a central role in our reasoning about and with concepts, (b) examine socially relevant stereotypes, and (c) investigate the structure of concepts.
This content is subject to copyright. Terms and conditions apply.
Synthese (2024) 204:39
https://doi.org/10.1007/s11229-024-04669-x
ORIGINAL RESEARCH
Salient semantics
Kevin Reuter1
Received: 17 November 2022 / Accepted: 6 June 2024 / Published online: 17 July 2024
© The Author(s) 2024
Abstract
Semantic features are components of concepts. In philosophy, there is a predominant
focus on those features that are necessary (and jointly sufficient) for the application
of a concept. Consequently, the method of cases has been the paradigm tool among
philosophers, including experimental philosophers. However, whether a feature is
salient is often far more important for cognitive processes like memory, categoriza-
tion, recognition and even decision-making than whether it is necessary. The primary
objective of this paper is to emphasize the significance of researching salient features
of concepts. I thereby advocate the use of semantic feature production tasks, which
not only enable researchers to determine whether a feature is salient, but also provide
a complementary method for studying ordinary language use. I will discuss empirical
data on three concepts, conspiracy theory,female/male professor, and life,
to illustrate that semantic feature production tasks can help philosophers (a) identify
those salient features that play a central role in our reasoning about and with concepts,
(b) examine socially relevant stereotypes, and (c) investigate the structure of concepts.
Keywords Salience ·Semantic features of concepts ·Semantic feature production
tasks ·Necessity ·Conceptual analysis ·Conceptual engineering ·Theories of
concepts
1 Theoretical background
1.1 What are salient features?
According to most psychological and philosophical theories of concepts, the majority
of concepts can be analyzed into sets of features or components. Those features have
different attributes.1Advocates of classical theories of concepts are mostly interested
1In this paper I use the term ‘features’ to denote components of concepts, whereas the term ‘properties’
is reserved for aspects of objects. The term ‘attributes’ is used to refer to the logical status of features, e.g.,
necessary, sufficient, typical, salient.
BKevin Reuter
kevin.reuter@philos.uzh.ch
1Institute of Philosophy, University of Zurich, Zürichbergstrasse 43, 8044 Zurich, Switzerland
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 2 of 20 Synthese (2024) 204 :39
in whether a feature has the attribute of necessity, i.e., whether the feature is a necessary
component of a concept. Advocates of prototype and exemplar theories care more
about whether a feature is typical or prototypical, i.e., whether most objects that fall
under a concept have the property referred to by that feature, or whether a particularly
noteworthy exemplar has that feature.
Necessity and typicality are not the only attributes that play an important role in our
representation of kinds. A feature might be universal, i.e., all objects falling under a
concept have the property referred to by that feature, but not necessarily so. A feature
might also be salient, which means—very roughly—that some objects falling under
a concept have a property referred to by that feature that is striking, i.e., it stands out
from other properties in our representation of the kind.2
Here is an example to illustrate differences between those attributes. The concept of
shark might be characterized to have the following set of features: <is a fish>,<has
5–7 pairs of gills>,<is predatory>,<attacks humans>. Sharks are necessarily fish.
<is a fish>is thus a necessary feature of shark. All sharks have 5–7 pairs of gills,
but sharks might evolve to have more than 7 pairs of gills. The feature <has 5–7 pairs
of gills>is thus universal without being necessary. Most sharks are predatory, but not
all are. <is predatory>is thus only a typical, but not a universal or even necessary
feature of shark. And hardly any sharks attack humans, but <attacks humans>is
a highly salient feature, most likely because the potential danger of sharks plays an
important role in people’s reasoning about sharks.
Although I have provided a tentative characterization of salience above and have
discussed an example of a salient feature of a concept, readers might still be unclear
about what exactly is meant by salience. Unfortunately, there is no accepted definition
of salience. Instead, researchers seem to either
provide paraphrases that are likely to be uninformative and possibly circular (Slo-
man et al., 1998, my own characterization above);
state technical definitions (Del Pinal & Spaulding, 2018; Fischer & Engelhardt,
2020; Sloman et al., 1998);
operationalize instead of provide a definition (McRae et al., 2005).
Let me quickly take these three approaches in turn. First, stating that features are
salient if and only if the property they refer to is striking (see my characterization
above) seems to only trade one word for a synonymous term. More worryingly, if
one is pressed to say what it is for a property to be striking, one is easily led into a
circle by saying that a striking property is salient in our representation of the kind.
Sloman et al.’s (1998) claim that salient features are those that are ‘prominent’ in our
representation does not fare any better, unless an informative definition of ‘prominent’
is given, which is at least absent in Sloman et al.’s discussion.
Second, Sloman et al. also provide a more technical definition of salience:
Salience refers to the intensity of a feature, the extent to which it presents a
high amplitude signal in relation to background noise, in a way that is fairly
2Other attributes of features like diagnostic central,andsufficient do not play an important role for the
purposes of this paper.
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Synthese (2024) 204 :39 Page 3 of 20 39
independent of context. For example, the brightness of a bright light or the
redness of a fire engine are salient features. (1998, p. 193)
It seems that Sloman et al. take the analogy with perceptual salience quite literally.
However, while perceptual salience of a stimulus and the salience of a feature of a
concept might share some commonalities, the analogy breaks down fairly quickly
when we consider functional and abstract features. For example, it is hard to tell what
the signal-to-noise ratio for the feature <attacks humans>is supposed to be, or how
to cash out the salience of features for abstract concepts like conspiracy theory or
life in terms of its intensity or high amplitude signal. In a range of projects, Fischer
and Engelhardt investigate reasoning processes that are influenced by what they call a
linguistic salience bias. While their focus is more strongly on stereotypical inferences
due to dominant uses of words (see e.g., Fischer & Engelhardt, 2016, Fischer &
Engelhardt, 2019, see also Fischer & Sytsma, 2021), they also provide an extended
characterization of salience:
Salience (in this sense) is a function of exposure frequency, that is, of how often
the language user encounters the word in this sense. It is further modulated
by prototypicality (Rosch, 1978), where a sense of a polysemous word (e.g.,
“see”) is more or less prototypical depending upon whether it stands for more
or less prototypical examples of the relevant category (e.g., more or less proto-
typical cases of seeing). The more salient a use is for a hearer, the more rapidly
and strongly the situation schema associated with it gets activated. (Fischer &
Engelhardt, 2020, p. 418)
Cashing out salience as a function of frequency and prototypicality seems very
promising, especially given the results from empirical studies (see below). A pos-
sible downside of this characterization is that prototypicality is usually considered to
be itself a function of frequency, and thus, we would need to know more about Fischer
& Engelhardt’s definition of prototypicality. Additionally, their work focuses mostly
on polysemous uses of terms, for which we can expect other factors to play a more
important role for salience.
Other researchers (e.g., McRae et al., 2005) take the salience of a feature to be
what is revealed by certain experimental tasks, e.g., memory retrieval tasks. How-
ever, this approach rather operationalizes the concept of salience but does not provide
an independent definition. Consequently, we cannot tell whether a feature is salient
because it is quickly retrieved from memory or whether a feature is quickly retrieved
from memory because it is salient. With this not entirely satisfying state of affairs,
let us consider the role that salience plays for traditional philosophers, experimental
philosophers and psychologists.
1.2 (Experimental) Philosophers and psychologists
In the practice of conceptual analysis, philosophers typically focus on identifying
components that are necessary for applying a concept. This approach often leads them
to adopt the classical theory of concepts. According to this theory, concepts are under-
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 4 of 20 Synthese (2024) 204 :39
stood as collections of features that are both necessary and jointly sufficient for their
application.3,4The dominant method of conceptual analysis for determining the nec-
essary and jointly sufficient features of a concept is the method of cases: Philosophers
devise thought experiments that constitute possible or actual counterexamples to the
proposed definition of a concept.
Experimental philosophers have challenged the idea that a small number of experts,
i.e., professional philosophers, can reliably and robustly reveal whether a putative
counterexample works or whether it fails (Brun & Reuter, 2022; Mallon et al., 2009).
Most experimental philosophers, however, rarely challenge the predominance of the
search for necessary and jointly sufficient features of concepts. In other words, the
method of cases is also a prevalent approach among experimental philosophers. While
vignette studies do not need to be designed to test for necessary and jointly suffi-
cient features, they often are.5Why have philosophers been less intrigued by other
attributes of features, especially by the attribute of salience? Arguably, the issue of
feature salience has been largely neglected in philosophical debates because philoso-
phers commonly hold that their main objective is to probe deeper than the obvious
or ‘salient’ aspects that are immediately accessible to most individuals. Instead, their
goal is to venture past these initial impressions to uncover the more intricate and pro-
found essence of various phenomena. However, this paper argues that disregarding
the salience of features as irrelevant to philosophical inquiry is a significant over-
sight. It will demonstrate that acknowledging and examining the salience of features
is not just beneficial but essential for a comprehensive and accurate understanding of
various concepts, offering a richer and more complete perspective in philosophical
explorations.6
Several scholars, including Machery (2017) and Isaac (2021a,2021b), have under-
scored the importance of adopting psychological approaches in analyzing concepts and
highlighting areas for conceptual change. Despite this, the influence of feature salience
on our engagement with and understanding of philosophical concepts remains largely
uncharted territory. A notable exception can be found in the works of Fischer and his
colleagues, who investigate the effects of what they term “linguistic salience bias” on
reasoning processes, as seen in Fischer and Engelhardt (2016,2020), and Fischer and
Sytsma (2021). Aside from Fischer and colleagues’ work on the linguistic salience
3Sometimes, especially in discussions on natural kinds, the theory–theory of concepts is considered as an
alternative to the classical theory of concepts. However, even then, the essential features of a natural kind
concept are the ones that interest philosophers most, and not so much the superficial features that might be
highly salient.
4For a discussion on the significance of the classical theory of concepts in philosophy, see especially
Laurence and Margolis (1999).
5Other methods like the cancellation paradigm (Willemsen & Reuter, 2021) corpus analysis (e.g. Hansen
et al., 2021;Sytsmaetal.,2019), eye tracking (e.g., Fischer et al., 2021), and statistical learning (Nichols,
2021) have entered the experimental–philosophical stage, which are usually not utilized to test for necessary
features of concepts.
6Recently, the philosophical study of salience has received some attention (see e.g., Archer, 2022; Michael-
son & Nowak, 2022; Ratcliffe & Broome, 2022; Watzl, 2022; Whiteley, 2022). However, note that while
these philosophers, as well as many psychologists and linguists, are focused on the salience of a stimulus
that can capture people’s attention (Wu, 2011), e.g., a red fire hydrant on a sidewalk, this paper is focused
on the salience of features of a concept (Breheny, 2008; Fischer & Sytsma, 2021; Fischer et al., 2021).
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Synthese (2024) 204 :39 Page 5 of 20 39
bias, the importance of the salience of a feature of a concept has been noted in the study
of generics, see, in particular, Leslie (2008). According to Leslie’s view, if a particular
feature (like <attacks humans>) is highly salient in a concept like shark, people are
likely to formulate a generic statement based on that feature (“Sharks attack humans”),
even if the characteristic is not statistically prevalent in the category of sharks. Despite
the significance of feature salience for the study of generics, little empirical work has
in fact been conducted to determine the salient features of concepts.
Psychologists, in contrast to both traditional and experimental philosophers, are less
impressed by the classical theory of concepts and the search for necessary and jointly
sufficient features. Not only are there surprisingly few success stories of the classical
theory (Laurence & Margolis, 1999), many studies have shown that our categorization
and reasoning processes are often best explained by prototypical representations and
encoded exemplars (Rosch, 1978; Hampton, 1995). Whether or not the classical theory
can somehow accommodate this research is a matter of ongoing debate (see e.g.,
Lakoff, 2007; Machery, 2009). I am not taking any sides in this discussion. What can
be said though with certainty is that psychologists take very seriously and investigate
thoroughly the salience, typicality, centrality, and diagnosticity of features, whereas
philosophers consider those attributes of a concept’s features to a much smaller extent.
Despite psychologists’ interest in salient and typical features of concepts, there
is relatively little work on the salient and typical features of individual concepts,
especially for philosophically relevant concepts. As an example, take the concept of lie
for which philosophers and psychologists have advanced our knowledge hand-in-hand.
The case of lie underscores my claim that even experimental philosophers are usually
strongly invested in the traditional program of finding necessary and jointly sufficient
conditions. Coleman and Kay (1981) empirically studied and developed a prototype
semantics for the concept lie. Experimental philosophers have contributed widely to
the literature on lie within the last 10 years. Although some of this research belongs
both philosophically and methodologically to the best of experimental philosophy,
those experimental philosophers have, unfortunately in my opinion, ‘gone back’ to
frame their results in terms of necessary and sufficient conditions (Rutschmann &
Wiegmann, 2017; Turri & Turri, 2015; Wiegmann & Willemsen, 2017).7
Thus, on the one hand, we have scholars—the psychologists—who are interested
in typical and salient features, but mostly in order to understand how people reason
with and about concepts more generally. On the other hand, we have scholars—the
philosophers—who are interested in individual concepts like truth,lie,knowledge,
conspiracy theory, etc., but then mostly attend to those features of these concepts
that might turn out to be necessary for their application.
7Rutschmann and Wiegmann state that “[o]ur results indicate that the intention to deceiveis not a necessary
condition for lying.” (2017, p. 438). Turri and Turri claim that “[o]ur main conclusion is that, contrary to
the standard view, falsity is a necessary component of lying and, thus, that lying has an important non-
psychological element.” (2015, p. 162) and Wiegmann and Willemsen hold that “[i]n this paper, we showed
that for something to be a lie, subjective falsity at the semantic level is not necessary.” (2017, p. 3521)
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 6 of 20 Synthese (2024) 204 :39
1.3 The semantic feature production task
The method of cases is not an uncontroversial method, but it surely is the dominant
paradigm to explore which features of a concept are considered to be necessary. What
about salient features? How can we explore which features are the salient features
of a concept? The probably most widely used method to determine salient features
of concepts is the so-called semantic feature production task, also known as feature
listing task (for some early examples, see e.g., Hampton, 1979; Barsalou, 1983;for
an in-depth discussion see, e.g., Machery, 2017).8Interestingly, there is no prescribed
way of conducting a semantic feature production task. So, let’s look more closely
at how one of the most influential papers (McRae et al., 2005) on semantic feature
production tasks goes about doing this (see also Wu & Barsalou, 2009). This is what
they presented their participants with:
We want to know how people read words for meaning. Please fill in features of
the word that you can think of. Examples of different types of features would
be: how it looks, sounds, smells, feels, or tastes; what it is made of; what it is
used for; and where it comes from. Here is an example:
duck: is a bird, is an animal, waddles, flies, migrates, lays eggs, quacks, swims,
has wings, has a beak, has webbed feet, has feathers, lives in ponds, lives in
water, hunted by people, is edible
Complete this questionnaire reasonably quickly, but try to list at least a few
properties for each word. (McRae et al., 2005)
After participants were presented with these instructions, they were then simply given
a list of words and fields to fill in features that came to mind. These features largely
fell into four categories: sensory (e.g., has fur), functional (e.g., you can sit on it),
encyclopedic (e.g., lives in woods), and taxonomic (e.g., is a vegetable).
The results of their semantic feature production tasks for 571 items allows for two
important observations. First, typicality is arguably the single most important predictor
for the frequency with which a feature is named, where typicality is understood as the
frequency with which members of a certain category possess the property referred to
by the feature.9To illustrate, consider the case of a knife: Not all knifes are sharp,
are used for cutting, and have a handle. But many, and perhaps even most, knifes are.
Thus, being sharp, used for cutting and having a handle are highly frequent properties
of knifes. Thus, rather unsurprisingly, <sharp>,<used for cutting>,<has a handle>
are typical features and among the most common features named in a semantic feature
production task for knive.
8Other methods that have been used to identify salient features of concepts are sentence completion tasks
(e.g., Fischer & Engelhardt, 2020) as well as non-literal comparisons, see e.g., Barsalou (1982)andOrtony
et al. (1985).
9It might be argued that the example of a duck that was given to participants biased people into giving
responses that are typical features of concepts. However, similar outcomes have been observed in other
methodologies (see e.g., Hampton, 1979), and the results presented in this paper suggest that the example
was not biasing people into giving typical features.
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Synthese (2024) 204 :39 Page 7 of 20 39
Second, “participants’ responses are somewhat biased toward information that dis-
tinguishes among concepts—that is, the pieces of information that enable people
to distinguish a concept from other, similar concepts” (2005, p. 549). For example,
take the feature <attacks humans>of the concept shark. Sharks hardly ever attack
humans. However, other fish are even more unlikely to attack humans. Thus, among
all fish, we can distinguish sharks from other fish easily by their propensity to attack
humans, even if the propensity is very low. On the flip side, people are very unlikely
to state the feature <has a kidney>for the concept human, although all humans
have a kidney. <Has a kidney>simply does not allow us to distinguish humans from
many other animals. Rosch (1978) made popular the term ‘cue validity’ to refer to
the conditional probability of an object falling in a particular category given a par-
ticular property. Cue validity is greater the more the feature is considered to apply to
members of the category in question, and the less the feature is considered to apply to
members of other categories. Thus, <attacks humans>has high cue validity because
we associate the property of attacking humans strongly with sharks and with hardly
any other fish. And <has a kidney>has relatively low cue validity despite it being
universal for humans, because we also associate the property of having a kidney with
many other animals.
The more typical and the more cue valid a feature is, the more likely it will be
stated frequently in semantic feature production tasks. If we take semantic feature
production tasks to reveal some of the most salient features of concepts, then typical-
ity and cue validity seem to be the two most important predictors for salience. Features
derived from semantic feature production tasks have been shown to be crucial for cog-
nitive processes like memory, categorization, recognition and even decision-making
(Ashcraft, 1978; Cree et al., 1999; Hampton, 1979; Smith et al., 1988; Solomon &
Barsalou, 2001; Vigliocco et al., 2004). Thus, the importance of salient features of
concepts for various cognitive processes can hardly be overstated.
1.4 Proof of concept
So far, I have introduced the notion of salience (Sect. 1.1). I have then argued that
whereas philosophers are interested in individual concepts but rarely in the salient fea-
tures of those concepts, psychologists are interested in the salient features of concepts,
but rarely in any individual concepts that are also relevant for philosophers (Sect. 1.2.).
Lastly, I discussed the semantic feature production task as one of the primary methods
to reveal the salient features of concepts (Sect. 1.3).
Perhaps, salient features of concepts have not been discussed very much by philoso-
phers, because salient features are simply not particularly interesting when individual
concepts are at stake. Thus, the burden of proof is certainly on those like me who
argue that we should care about whether a feature of a concept is salient. In the next
section, I will therefore go through three empirical studies to try to make the point that
we have been wrong in neglecting salient semantics.
Before proceeding, it is important to address a further theoretical question: Should
the study of salient features be classified within the domain of semantics? Using the
term ‘salient semantics’ indeed marks a departure from traditional truth-conditional
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 8 of 20 Synthese (2024) 204 :39
semantics. Nonetheless, there are several compelling reasons why ‘salient semantics’
is an appropriate term. First, this approach resonates with Putnam’s (1970) sugges-
tion that prototype structures are an integral part of a term’s meaning, even though
they don’t directly determine the word’s reference. Second, the exploration of salient
features is fundamentally different from pragmatic analysis. While pragmatics deals
with the use of language in context and the implications of that use, the study of
salient features focuses on the inherent characteristics of concepts as they are under-
stood independently of specific contexts. Semantic feature production tasks typically
involve collecting features of concepts in a context-independent manner, i.e., partici-
pants are asked to name features of concepts without having first read a vignette, or
having been primed about a specific subject. Third, semantics is fundamentally con-
cerned with the meaning and interpretation of words and phrases in language. Salient
features of concepts play a crucial role in how we understand and ascribe meaning to
various terms and concepts. By examining these features, we gain insights into how
meanings are constructed, interpreted, and conveyed in language. Or so I hope to show
in the next section.
2 Empirical studies
In this section of this paper, I aim to demonstrate how an analysis of the salient features
of concepts enables us to achieve three key objectives: (a) pinpoint the features that
are crucial in our reasoning about and with concepts, (b) scrutinize socially pertinent
stereotypes, and (c) explore the intrinsic structure of concepts. To illustrate these
points, I will engage in detailed discussions of three recent empirical studies, each
focusing on a different concept: conspiracy theory,female/male professor,
and life. Through these case studies, we will see how an investigation into the salient
features not only enriches our understanding of these specific concepts but also offers
broader insights into the dynamics of conceptual analysis.
2.1 Study 1: When salience trumps necessity
There are many exciting philosophical projects which aim at revealing the necessary
features of philosophically interesting concepts. My aim is not to undermine these
efforts, which are integral to mainstream analytic philosophy. My overall point is
different: understanding the salient features of concepts is crucial, as it enables us
to address several philosophically important questions. Therefore, research that aims
to identify necessary features and research targeting salient features should not only
coexist but also enrich each other. That said, there are concepts, where it seems that
salience trumps necessity. Let’s take a look at such a concept.
2.1.1 The case of conspiracy theory
The prevailing view among analytic philosophers is that conspiracy theories are
fundamentally theories about conspiracies. This viewpoint is supported by a range
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Synthese (2024) 204 :39 Page 9 of 20 39
of scholars, including Basham and Dentith (2016), Cassam (2019), Coady (2003),
Cohnitz (2018), Feldman (2011), Harris (2018), Keeley (1999), Pigden (2007), and
Räikkä (2018), although it is important to note that these scholars don’t necessarily
agree on a singular definition of ‘conspiracy theory’. According to this dominant view,
conspiracy theory is not a negative evaluative concept but rather seen as a descrip-
tive concept. This implies that features like <deficient>,<crazy>,orsimply<bad>
are not necessary features of conspiracy theory. While some philosophers recog-
nize that conspiracy theories are often perceived negatively, such evaluative aspects
are considered to be at most pragmatic features (Pigden, 2007).
One argument for proposing a descriptive account of conspiracy theory is
straightforward: by applying the method of cases, philosophers can illustrate that cer-
tain theories are identified as conspiracy theories without evaluating them negatively.
Take, for instance, the theories surrounding the Watergate scandal. These are seemingly
aptly categorized as conspiracy theories, given Nixon’s involvement in the conspiracy
and subsequent cover-up. Yet, such theories are not regarded as epistemically flawed,
irrational, or morally reprehensible. Hence, it’s reasonable to consider the Watergate
scandal as a compelling counterexample to the notion that the term ‘conspiracy theory’
is inherently evaluative.10
Even though this putative counterexample challenges the idea that <being
deficient>or a similar evaluative aspect is intrinsic to the concept conspiracy the-
ory, it is still intriguing to explore how salient negative evaluations factor into our
understanding of conspiracy theories. In this context, Napolitano and Reuter (2023)
conducted a study where they gathered people’s responses using a modified semantic
feature production task. Their findings indicate that negative evaluative elements are
indeed salient in our representation of conspiracy theories. However, their approach
diverged from the standard methodology; they asked participants to identify features
they deemed necessary for a concept to qualify as a conspiracy theory. To build on
this, I implemented another semantic feature production task, with a straightforward
prompt: “Please tell us: Which features are characteristic of a conspiracy theory?” In
this task, 40 participants were given three fields to input three features, aiming to gain
further insight into the salient features of conspiracy theory.
In this study, one participant listed only examples of conspiracy theories, such
as ‘flat earth’, ‘hollow earth’, and ‘lizard’, rather than identifying features. Among
the remaining 39 participants, 34 participants (which constitutes 87% of the sample)
included at least one negative evaluative term in their response. These terms ranged
from ‘ambiguous evidence’ to descriptors like ‘far-fetched’, ‘confusing’, ‘misleading’,
‘outlandish’, ‘self-importance’, ‘gossip’, ‘arrogance’, and ‘lies’, indicating that nega-
tive evaluation is a highly salient aspect in how people conceive of conspiracy theories.
The responses of the first 15 participants, as shown in Table 1, further illustrate this
trend, providing a detailed view of how these theories are commonly characterized.11
10 Some of Napolitano and Reuter (2023)’s results put pressure on the idea that many people wouldconsider
the Watergate scandal a conspiracy theory.
11 Through their corpus-analytic studies, Napolitano and Reuter (2023) and Reuter and Baumgartner (2024,
forthcoming) have provided evidence that the term ‘conspiracy theory’ is commonly used with negative
implications. This research leads to an intriguing and largely unexplored question: How effectively can
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 10 of 20 Synthese (2024) 204 :39
Table 1 Responses of the first 15 participants in the semantic feature production task
Participant 1st Term 2nd Term 3rd Term
Person 1 Hopefulness Arrogance Over confidence
Person 2 Lies Assumptions Belief
Person 3 Unlikely outcomes Unable to pinpoint sources Sensationalism
Person 4 Not educated Paranoid Poor
Person 5 Unlikely Far-fetched Unbelievable
Person 7 Paranoia Theories Falsehoods
Person 8 Little evidence Somewhat mysterious Extreme ideas
Person 9 Technology Government involvement Allegation of suppression of info
Person 10 Lies Fear Distrust
Person 11 Made up Sensational Far fetched
Person 12 Sinister Motivates Going against expert analysis Paranoia
Person 13 Wild thoughts Terrible sources Gullible people
Person 14 Paranoia Suspicion Deception
Person 15 Usually false Outlandish The government are behind everything
2.1.2 Discussion
The results of the semantic feature production task on conspiracy theory demon-
strate that an overwhelming majority think that negative features like <far-fetched>
and <outlandish>are characteristic features of conspiracy theory. Thus, taking
semantic feature production tasks to be in the business of revealing salient features, a
highly salient feature in our representation of conspiracy theories is something akin to
<is epistemically deficient>. However, let us not forget, that the concept conspiracy
theory does not seem to necessarily be an evaluatively negative concept. The case
of the Watergate scandal provides an intuitively compelling example that conspiracy
theories can be true, justified and rational theories. So, what shall we do with our
findings?
To truly understand and interpret people’s attitudes towards conspiracy theories, it
is important to grasp the salient features of conspiracy theory. Similarly, com-
prehending how individuals reason with the term ‘conspiracy theory’ requires an
understanding of these salient features. For the purpose of conceptual engineering,
particularly in relation to everyday understanding of ‘conspiracy theory’, identify-
ing these salient features is crucial. Simply analyzing specific instances, such as the
Watergate scandal, falls short in providing the necessary insights for these explo-
rations. Consequently, being aware of the necessary features required to apply the
term ‘conspiracy theory’ has limited utility in offering constructive responses.
corpus analysis uncover the salient features of a concept? As it stands, the extent to which this method can
yield comprehensive insights into concept characterization remains an open area for investigation.
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Synthese (2024) 204 :39 Page 11 of 20 39
2.2 Study 2: On stereotypes
Exploring salient features not only provides insights into how people reason and
understand various concepts, but it is also crucial for understanding how certain stereo-
types take shape. Both psychological and philosophical research extensively examine
aspects such as (a) identifying prevalent stereotypes and biases, (b) tracing their ori-
gins, (c) analyzing their ethical implications, and (d) exploring potential interventions.
Philosophers, in particular, are adept at addressing the ethical consequences of stereo-
types and biases within the context of broader societal injustices and specific instances
of discrimination. Additionally, philosophers across various disciplines are trained to
critically assess the vehicles of thought, i.e., the concepts and language we employ
to describe and think about the world. While modifying the language and concepts
we use might not always be the most direct or effective method to combat harmful
stereotypes, the prevailing consensus is increasingly acknowledging that our words
and concepts are not neutral and require careful consideration and, potentially, change.
Stereotypical thinking often arises from the salient features of the concepts we hold,
a phenomenon also highlighted in the research by Fischer and Engelhardt (2020). It
is common for these stereotypes to manifest in beliefs* such as Asians excelling in
mathematics or male Italians being sexist. However, these beliefs, as indicated by the
asterisk, need not be explicitly endorsed; they are often held implicitly, as discussed
by Holroyd et al. (2017) and Schwitzgebel (2010). Importantly, these stereotypical
features are not viewed as necessary for the application of a concept. For instance, no
one genuinely believes that being proficient in math is a necessary characteristic of the
concept Asian person. Rather, such stereotypes reflect the traits that are conceived
of as salient within certain groups, like Asians or male Italians. These conceptions
highlight how stereotypes inform our understanding of different social groups. The
study presented here emphasizes the critical importance of examining salient features
in discussions surrounding stereotypes.
2.2.1 The case of female and male professor
In an influential article, Leslie et al. (2015) show that women are under-represented in
fields were brilliance is believed to be a more important determinant of success than
hard work. They also provide an explanation for this finding. They hypothesize that
“women are stereotyped as not possessing such talent.” (2015, p. 262). This hypothesis
would indeed explain the empirical finding that women are under-represented in fields
such as mathematics, physics, etc.
Surprisingly, despite the availability of experimental methods, Leslie et al.’s hypoth-
esis hasn’t been directly tested. To address this, Del Pinal et al. (2017) conducted a
semantic feature production task to identify the most salient features associated with
the concepts of female professor and male professor. If Leslie and colleagues’
assertion is accurate—that women are stereotypically believed to lack brilliance—then
this belief should be reflected in the salient features identified for female professor.
This approach offers a direct empirical test of the hypothesis, seeking evidence for the
stereotype in question.
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 12 of 20 Synthese (2024) 204 :39
Table 2 Results of the semantic feature production task in Del Pinal et al. (2017)
Male (%) Female (%) χ2p
Professor + hardworking 21.3 39.5 4.71 0.030
Professor + smart 72.4 76.6 1.26 0.473
Actor/actress + hardworking 7.5 10.9 0.36 0.547
Actor/actress + smart 17.0 10.9 0.77 0.380
Baker + hardworking 24.5 19.2 0.41 0.522
Baker + smart 10.2 1.9 3.09 0.079
A total of 312 participants were recruited via Amazon Mechanical Turk. Participants
were requested to write down features for specific social categories. They were ran-
domly assigned either to one of the two test conditions (female professor or male
professor), or to one of four control conditions (female baker,male baker,
female actress,female actor). The target stimuli were as follows:
Imagine that Mary/Jack is a professor at a university [an actor/actress; a baker].
Please list five features that you think are typical of Mary/Jack.
(from Del Pinal et al., 2017)
The results are displayed in Table 2and reveal two noteworthy findings: First,
contrary to the hypothesis that female professors are stereotypically viewed as less
brilliant, the participants in this semantic feature production task attributed terms such
as ‘smart’ and ‘intelligent’ equally to both female and male professors. This finding
provides little support for the explanation proposed by Leslie and colleagues regarding
the stereotype of female professors’ brilliance. Second, although there was no signifi-
cant difference in the frequency of terms like ‘smart’ and ‘intelligent’ being associated
with both genders, a notable difference emerged in the proportion of participants who
used terms synonymous with ‘hardworking’. Intriguingly, the term ‘hardworking’
and its synonyms were almost twice as likely to be associated with female professors
compared to male professors.
In a subsequent study, Del Pinal and colleagues offered an empirically supported
alternative explanation for the perceived association between gender and brilliance.
Their study focused on exploring gender-specific associations between being ‘smart’
and ‘hardworking’. Participants were asked to estimate the number of hours per week
a female and a male professor would need to work. The findings were telling: when
the smartness of female professors was emphasized, they were perceived as needing
to work more hours compared to the control condition. In contrast, this increase in
perceived work hours was not observed for male professors when their smartness
was highlighted. Overall, the research by Del Pinal et al. suggests that the stereotype
linking brilliance and gender exists within the dependency networks of the concepts
female professor and male professor, but this stereotype is not evident at the
level of feature salience.
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Synthese (2024) 204 :39 Page 13 of 20 39
2.2.2 Discussion
Semantic feature production tasks not only enable the identification of a concept’s
most salient features (as demonstrated in Study 1), they also provide insights into
the extent and nature of stereotypical thought. As such, these tasks offer a relatively
efficient and direct method for examining various hypotheses related to the stereotypes
held by people. In the specific instance of Del Pinal et al.’s study, this approach was
instrumental in casting doubt on Leslie’s hypothesis that women are stereotypically
perceived as having less intelligence.
2.3 Study 3: On the structure of concepts
In the previous two subsections, we have seen that detecting the salient features of
concepts allows us (a) to identify those features that are likely to dominate people’s
thinking in various cognitive processes, and (b) to examine those salient features that
are likely to play a crucial role in stereotypical thinking. In the following study (Study
3), I aim to show that investigating salient features through semantic feature production
tasks can provide us with insights not only into the meaning of concepts but also into
the structure of concepts.
2.3.1 The case of LIFE
The concept life has been highlighted to be particularly resistant to a (classical) def-
inition that receives widespread agreement. Chyba and McDonald (1995, p. 216), for
example, claim that “it is now a commonplace that the various proposed definitions [of
life] virtually all fail”. While Chyba and McDonald might be right in their assessment,
they do not provide an answer of why that is the case. In trying to make sense of the
state of confusion, drawing on Wittgenstein’s notion of family resemblance, Pennock
(2012, p. 5) claims that “life is a cluster concept with fuzzy boundaries”. However,
conclusions in regards to the structure of the concept life seem to be at best premature
without any empirical evidence on the matter.
Beisbart and Reuter (2021) presented laypeople with a semantic feature production
task. More specifically, they asked 102 participants (65 female, 36 male, 1 non-
identied), to write down up to three answers to one of the following two different
versions of a semantic feature production task:
First version. “Which features are characteristic of species of living beings? You
can name up to three features.”
Second version. “Which features do you think distinguish species of living beings
from non-living entities?”
The answers of the first 15 participants are displayed in Table 3. The most fre-
quently named responses fell into the categories <growth>(47%), <breathing>
(46%), <reproduction>(35%) and <nutrition>(31%). People gave responses indica-
tive of what “living beings do at the level of a whole living being, and most of these
features are observable for many life forms” (2021). People hardly gave answers that
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 14 of 20 Synthese (2024) 204 :39
Table 3 Responses of the first 15 participants to the semantic feature task
Beisbart and Reuter used eight categories to classify the responses: breathing cells (no response among the
first 15 participants), evolution, growth, movement, nutrition, perception/consciousness, reproduction. The
cells left white indicate responses that were not categorized. Table taken from Beisbart and Reuter (2022)
were classified to refer to the material (organic matter) or structure of the underlying
material (cells (10%)).
In a second study, Beisbart and Reuter aimed to find out which features of living
beings people consider to be universal, i.e., which features are thought to hold for all
species of life. In order to examine which features people consider to be universal,
they asked them “What percentage of species consist of living beings that [feature]?”
People’s answers were measured on a scale ranging from 0 to 100% in steps of 1%. In
contrast to the semantic feature production task, an entirely different outcome emerged.
Whereas only 10% of the participants named <cells>or <material>in the semantic
feature production task, a whopping 68% thought that 100% of species of living beings
are made of organic material, and 64% considered all species to be made of cells. In
contrast, <growth>and <nutrition>received much lower numbers.
2.3.2 Discussion
The research conducted by Beisbart and Reuter indicates that focusing solely on nec-
essary features overlooks critical elements of the concept of life. A key finding from
their empirical studies is the distinction between salient features on the one hand,
and universal features on the other. This raises the question: How can we interpret
these differences? Beisbart and Reuter suggest that this division mirrors the inherent
structure of the concept of life, which they believe is a natural kind concept embody-
ing both an essence and observable surface properties. Their proposal outlines three
aspects: First, they posit that life is conceived of as a natural kind, underpinned by
an esseence, such as cellular or organic composition. Second, they argue that people
identify this natural kind through salient macroscopic features, like <growth>and
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Synthese (2024) 204 :39 Page 15 of 20 39
<nutrition uptake>. Third, they contend that the essence can be identified by current
scientific knowledge.
Claims, according to which life cannot be defined (Machery, 2012), or that life
is a family resemblance concept (Pennock, 2012) were not based on empirical data
that took into account the importance of salient features. However, this case shows
that finding out about the structure of a concept often requires investigating the salient
features of a concept. The results of these investigations can then help us to capture
more precisely the semantics of our concepts as well as help developing more reliable
and empirically grounded theories.
3 General discussion
The identification of the salient features of concepts plays at most a minor role in
philosophical studies so far. Whenever philosophers are concerned with analyzing a
certain concept, they are likely to focus on what the necessary and jointly sufficient
features of a concept are. Experimental philosophers have certainly departed from
the obsession on necessary features in many philosophical projects. However, even
experimental philosophers seem to be strongly focusing on the traditional program of
uncovering necessary and sufficient conditions.
Psychologists have uncovered a range of cognitive processes where the salience
of features plays a more crucial role than their necessity for the application of a con-
cept. These processes encompass memory, recognition, categorization, and reasoning,
among others. While psychologists have given considerable attention to the salience
of features in concepts within their field, there’s a noticeable lack of focus on features
of concepts that hold philosophical significance. This oversight is understandable, as
concepts deemed important in philosophy may not always align with those considered
relevant in psychological studies. Therefore, the divergence in interest between these
two disciplines, particularly regarding concept features, is not surprising.
Despite the importance of salience for various cognitive processes, philosophers
might still be justified in disregarding research into the salient features of concepts.
Why is that? If our philosophical interests are only peripherally, if at all, affected by
the role that salient features play, then research time and mental effort are better put
into other projects. It seems, for example, that questions about how the salience of
features of concepts such as free will influence how those concepts are retrieved
from memory, are philosophically not overly exciting. The primary aim of this paper
was therefore to make a compelling case that philosophers should be interested in
salient semantics.
In the empirical part of this paper, I have presented studies that underline the impor-
tance of salient semantics for the concepts conspiracy theory,female professor,
and life.12 It seems to me that the results of the semantic feature production tasks are
indeed relevant for a range of philosophical debates, both at the level of those individ-
ual concepts as well as at the level of more general questions, like, how is evaluative
12 For further and very recent applications of the semantic feature production task for philosophical projects,
see Reuter (2019) on the concept of colleague Erbach et al. (2023)’s work on race, as well as Reuter,
Latham and Varga’s (ms) research on the concept of health.
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 16 of 20 Synthese (2024) 204 :39
content encoded in our concepts?, how do stereotypes work?, how are our concepts
structured?, etc. That said, there are a lot of issues and questions one can raise about
the role of the semantic feature production task. We still know relatively little about
the cognitive processes involved when retrieving features of abstract concepts in a
semantic feature production task. As most philosophically relevant concepts are fairly
abstract, it might well be that we cannot make reliable inferences from the results of
studies on concrete concepts (e.g., McRae et al., 2005; Vinson et al., 2008; Buchanan,
2019) to abstract concepts. Perhaps most worryingly, the lack of knowledge about the
cognitive processes involved in semantic feature production tasks for abstract con-
cepts, raise the very real possibility that the experiments tap into different properties
of these concepts.13
I would like to close by briefly considering the relevance of salient semantics for
two of the most central research methods in philosophy, conceptual engineering and
conceptual analysis.
3.1 Conceptual engineering
Conceptual Engineering has emerged as an umbrella term for explication and ame-
lioration. While explication projects aim to improve our concepts to make them more
fruitful for scientific purposes, ameliorative projects aim to improve our concepts (a)
for better public discourse and reasoning, (b) to eliminate sources of misunderstand-
ing and confusion, and (c) to reduce discrimination. I am certainly not the first to
argue that amelioration shouldn’t rely too much on a classical conception of concepts.
Other researchers (Machery, 2017: chap. 7; Fischer, 2020; Isaac, 2021a,2021b)have
highlighted the need for more “psychological” approaches both for identifying which
aspects of our concepts need improvement, and also for determining how new concepts
can be more successfully implemented. The results of the semantic feature production
tasks for conspiracy theory and female professor reinforce this claim. With-
out knowing how salient the evaluative content of conspiracy theory is, we do
not know, for example, the degree to which people may talk past each other when
discussing conspiracy theories, the degree to which the use of the term ‘conspiracy
theory’ has the potential to disparage certain theories and advocates of those theories,
etc. Furthermore, semantic feature production tasks are likely to allow insights into
how new or redefined concepts inherit unwanted features from related concepts. Thus,
when it comes to conceptual engineering, the need for salient semantics seems to be
immanently plausible.
3.2 Conceptual analysis
Conceptual analysis is traditionally conceived to be the process of (i) providing sets
of necessary and jointly sufficient features of concepts. Successful analyses are often
taken to be (ii) referentially invariant and (iii) feasibly performed by individual reflec-
tion on cases. There is a sense in which salient semantics has no bearing whatsoever
13 I owe these criticisms in part to Matteo Colombo (personal communication).
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Synthese (2024) 204 :39 Page 17 of 20 39
on conceptual analysis, given its focus on the necessity of features. That said, con-
ceptual analysis, as traditionally conceived, has received severe criticism. One strand
of criticism takes the underlying assumption of the classical theory of concepts to
be misguided (Chalmers & Jackson, 2001). A second strand takes experimental–
philosophical studies to show huge variation in the reference of concepts between
people on an individual level, as well as groups of people on a cultural level (Machery
et al., 2017; Reuter and Sytsma 2020; Weinberg et al. 2001). A third strand looks for
additional methods to circumvent various problems with the method of cases, e.g.,
corpus-analytic approaches (Andow, 2015; Fischer et al., 2015; Hansen et al., 2021;
Reuter, 2011; Sytsma et al., 2019). As a consequence of these objections, we find that
many philosophers entertain a looser concept of conceptual analysis that is not tied to
the classical theory of concepts, not tied to referentially invariant concepts, and not tied
to the method of cases. Although nowadays many philosophers don’t do conceptual
analysis as traditionally conceived, they, of course, still analyze concepts. Once we
adopt a wider and more liberal perspective on what it means to analyze concepts, there
is no good reason to exclude the investigation of salient features of concepts from the
philosophical task of analyzing concepts.
4 Conclusion
The primary objective of this paper was to emphasize the significance of research-
ing salient features of concepts. To underscore this point, I have detailed three case
studies, each exemplifying a unique context in which understanding salient features
is not just philosophically intriguing, but also essential for advancing knowledge in
specific areas. First, the case of conspiracy theory shows the importance of iden-
tifying salient features, which are pivotal in shaping our understanding and reasoning
about conspiracy theories. Second, examining the concept of female/male pro-
fessor highlights how salient features are instrumental in analyzing socially relevant
stereotypes. Finally, the study of life demonstrates the need of pinpointing salient
features to uncover the structure of concepts. While these three examples provide a
glimpse into the substantial philosophical utility of salient semantics, they represent
just a fraction of its potential applications and impacts.
Acknowledgements I am grateful to two anonymous reviewers for their detailed and insightful comments,
which significantly enhanced the manuscript. Earlier versions of this paper were presented at conferences in
Berlin, Lund, Prague, and Zurich, where I received invaluable feedback. I also wish to extend my thanks to
Richard Breheny,Matteo Colombo, Eugen Fischer, Ethan Landes, Edouard Machery, and PascaleWillemsen
for their engaging discussions on the material in this paper.
Funding Open access funding provided by University of Zurich The research of Kevin Reuter was funded
by the Swiss National Science Foundation (SNSF), grant number PCEFP1_181082.
Declarations
Conflict of interest The author has no conflict of interest to disclose
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 18 of 20 Synthese (2024) 204 :39
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://creativecommons.org/licenses/by/4.0/.
References
Andow, J. (2015). How “intuition” exploded. Metaphilosophy, 46(2), 189–212.
Archer, S. (2022). Salience: A philosophical inquiry. Routledge.
Ashcraft, M. (1978). Feature dominance and typicality effects in feature statement verification. Journal of
Verbal Learning & Verbal Behavior, 17, 155–164.
Barsalou, L. (1982). Context-independent and context-dependent information in concepts. Memory & Cog-
nition, 10(1), 82–93.
Barsalou, L. (1983). Ad hoc categories. Memory & Cognition, 10, 82–93.
Basham, L., & Dentith, M. R. X. (2016). Social science’s conspiracy theory panic: Now they want to cure
everyone. Social Epistemology Review and Reply Collective, 5(10), 12–19.
Beisbart, C., & Reuter, K. (2021). What Is the folk concept of life? Australasian Journal of Philosophy,
101, 1–22.
Breheny, R. (2008). A new look at the semantics and pragmatics of numerically quantified noun phrases.
Journal of Semantics, 25(2), 93–139.
Brun, G., & Reuter, K. (2022). Empirical studies on truth and the project of re-engineering truth. Pacific
Philosophical Quarterly, 103(3), 493–517.
Buchanan, E. M., Valentine, K. D., & Maxwell, N. P. (2019). English semantic feature production norms:
An extended database of 4436 concepts. Behavior Research Methods, 51(4), 1849–1863.
Cassam, Q. (2019). Conspiracy theories. Polity Press.
Chalmers, D., & Jackson, F. (2001). Conceptual analysis and reductive explanation. Philosophical Review,
110, 315–61.
Chyba, C., & McDonald, G. (1995). The origin of life in the solar system: Current issues. Annual Review
of Earth and Planetary Sciences, 23, 215–250.
Coady, D. (2003). Conspiracy theories and official stories. International Journal of Applied Philosophy,
17(2), 197–209.
Cohnitz, D. (2018). On the rationality of conspiracy theories. Croatian Journal of Philosophy, 18(2), 351–
365.
Coleman, L., & Kay, P. (1981). Prototype semantics: The English word lie. Language, 57(1), 26–44.
Cree, G. S., McRae, K., & McNorgan, C. (1999). An attractor model of lexical conceptual processing:
Simulating semantic priming. Cognitive Science, 23, 371–414.
Del Pinal, G., Madva, A., & Reuter, K. (2017). Stereotypes, conceptual centrality and gender bias: An
empirical investigation. Ratio, 30(4), 384–410.
Del Pinal, G., & Spaulding, S. (2018). Conceptual centrality and implicit bias. Mind & Language, 33(1),
95–111.
Erbach, K., Kenyah-Damptey, B., Berio, L., James, D., & Seyffarth, E. (2023). A comparative corpus study
of race and Rasse. Applied Corpus Linguistics, 3(1), 100044.
Feldman, S. (2011). Counterfact conspiracy theories. International Journal of Applied Philosophy, 21(1),
15–24.
Fischer, E. (2020). Conceptual control: On the feasibility of conceptual engineering. Inquiry.https://doi.
org/10.1080/0020174X.2020.1773309
Fischer, E., & Engelhardt, P. E. (2016). Intuitions’ linguistic sources: Stereotypes, intuitions, and illusions.
Mind & Language, 31, 67–103.
Fischer, E., & Engelhardt, P. E. (2019). Eyes as windows to minds: Psycholinguistics for experimental
philosophy. In E. Fischer & M. Curtis (Eds.), Methodological advances in experimental philosophy
(pp. 43–100). Bloomsbury.
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Synthese (2024) 204 :39 Page 19 of 20 39
Fischer, E., & Engelhardt, P. E. (2020). Lingering stereotypes: Salience bias in philosophical argument.
Mind & Language, 35(4), 415–439.
Fischer, E., Engelhardt, P., & Herbelot, A. (2015). Intuitions and illusions: From explanation and experiment
to assessment. In Fischer & Collins (Eds.), Experimental philosophy, rationalism, and naturalism (pp.
267–300). Routledge.
Fischer, E., Engelhardt, E., & Sytsma, J. (2021). Inappropriate stereotypical inferences? An adversarial
collaboration in experimental ordinary language philosophy. Synthese, 198(11), 10127–10168.
Fischer, E., & Sytsma, J. (2021). Zombie intuitions. Cognition, 215, 104807.
Hampton, J. (1979). Polymorphous concepts in semantic memory. Journal of Verbal Learning & Verbal
Behavior, 18, 441–461.
Hampton, J. (1995). Testing the prototype theory of concepts. Journal of Memory and Language, 34(5),
686–708.
Hansen, N., Porter, J. D., & Francis, K. (2021). A corpus study of “know”: On the verification of philoso-
phers’ frequency claims about language. Episteme, 18(2), 242–268.
Harris, K. (2018). What’s epistemically wrong with conspiracy theorising? Royal Institute of Philosophy
Supplement, 84, 235–257.
Holroyd, J., Scaife, R., & Stafford, T. (2017). What is implicit bias? Philosophy Compass, 12(10), e12437.
Isaac, M. G. (2021). Which concept of concept for conceptual engineering? Erkenntnis, 88, 1–25.
Isaac, M. G. (2021). Post-truth conceptual engineering? Inquiry, 67(1), 199–214.
Keeley, B. (1999). Of conspiracy theories. Journal of Philosophy, 96, 109–126.
Lakoff, G. (1987). Cognitive models and prototype theory. In U. Neisser (Ed.), Concepts and conceptual
development: Ecological and intellectual factors in categorization(pp. 63–100). Cambridge University
Press.
Laurence, S., & Margolis, E. (1999). Concepts and cognitive science. MIT Press.
Leslie, S.-J. (2008). Generics: Cognition and acquisition. Philosophical Review, 117(1), 1–47.
Leslie, S., Cimpian, A., Meyer, M., & Freeland, E. (2015). Expectations of brilliance underlie gender
distributions across academic disciplines. Science, 347(6219), 262–265.
Machery, E. (2009). Doing without concepts. Oxford University Press.
Machery, E. (2012). Why I stopped worrying about the definition of life... and why you should as well.
Synthese, 185(1), 145–164.
Machery, E. (2017). Philosophy within its proper bounds. Oxford University Press.
Machery, E., Stich, S., Rose, D., Chatterjee, A., Karasawa, K., Struchiner, N., Sirker, S., Usui, N., &
Hashimoto, T. (2017). Gettier across cultures. Noûs, 51(3), 645–664.
Mallon, R., Machery, E., Nichols, S., & Stich, S. (2009). Against arguments from reference. Philosophy
and Phenomenological Research, 79(2), 332–356.
McRae, K., Cree, G., Seidenberg, M., & McNorgan, C. (2005). Semantic feature production norms for a
large set of living and nonliving things. Behavior Research Methods, 37(4), 547–559.
Michaelson, E., & Nowak, E. (2022). On salience-based theories of demonstratives. In S. Archer (Ed.),
Salience: A philosophical inquiry (pp. 70–88). Routledge.
Napolitano, M. G., & Reuter, K. (2023). What is a conspiracy theory? Erkenntnis, 88(5), 2035–62.
Nichols, S. (2021). Rational rules: Towards a theory of moral learning. Oxford University Press.
Ortony, A., Vondruska, R. J., Foss, M. A., & Jones, L. E. (1985). Salience, similes, and the asymmetry of
similarity. Journal of Memory and Language, 24(5), 569–594.
Pennock, R. (2012). Negotiating boundaries in the definition of life: Wittgensteinian and Darwinian insights
on resolving conceptual border conflicts. Synthese, 185(1), 5–20.
Pigden, C. (2007). Conspiracy theories and the conventional wisdom. Episteme, 4(2), 219–232.
Putnam, H. (1970). Is semantics possible? Metaphilosophy, 1(3), 187–201.
Räikkä, J. (2018). Conspiracies and conspiracy theories. Argumenta, 3(2), 205–216.
Ratcliffe, M., & Broome, M. R. (2022). Beyond ‘salience’ and ‘affordance’: Understanding anomalous
experiences of significant possibilities. In S. Archer (Ed.), Salience (pp. 50–69). Routledge.
Reuter, K. (2011). Distinguishing the appearance from the reality of pain. Journal of Consciousness Studies,
18(9–10), 94–109.
Reuter, K., & Baumgartner, L. (2023). Conspiracy theories are not theories: Time to rename conspiracy
theories. In M. G. Isaac, S. Koch, & K. Scharp (Eds.), New perspectives on conceptual engineering.
Springer.
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
39 Page 20 of 20 Synthese (2024) 204 :39
Reuter, K., & Baumgartner, L., et al. (2024). Corpus analysis-building and using corpora: A case study
on the use of ‘conspiracy theory’. In A. Bauer, S. Kornmesser, L. Baumgartner, & K. Reuter (Eds.),
Experimental philosophy for beginners. Springer.
Reuter, K., & Sytsma, J. (2020). Unfelt pain. Synthese, 197(4), 1777–1801.
Reuter, K., Latham, A., & Varga, S. (ms). Health redefined: Lifestyle at the heart of modern health.
Rosch, E. (1978). Principles of categorization. Routledge.
Reuter, K., Löschke, J., & Betzler,M. (2019). What is a colleague? The descriptive and normative dimension
of a dual character concept. Philosophical Psychology, 33(7), 997–1017.
Rutschmann, R., & Wiegmann, A. (2017). No need for an intention to deceive? Challenging the traditional
definition of lying. Philosophical Psychology, 30(4), 438–457.
Schwitzgebel, E. (2010). Acting contrary to our professed beliefs or the gulf between occurrent judgment
and dispositional belief. Pacific Philosophical Quarterly, 91(4), 531–553.
Sloman, S., Love, B., & Ahn, W. (1998). Feature centrality and conceptual coherence. Cognitive Science,
22(2), 189–228.
Smith, E., Osherson, D., Rips, L., & Keane, M. (1988). Combining prototypes: A selective modification
model. Cognitive Science, 12, 485–527.
Solomon, K. O., & Barsalou, L. W. (2001). Representing properties locally. Cognitive Psychology, 43,
129–169.
Sytsma, J., Bluhm, R., Willemsen, P., & Reuter,K. (2019). Causal attributions and corpus analysis. In Fischer
& Curtis (Eds.), Methodological advances in experimental philosophy (pp. 209–238). Bloomsbury
Publishing.
Turri, A., & Turri, J. (2015). The truth about lying. Cognition, 138, 161–168.
Vigliocco, G., Vinson, D., Lewis, W., & Garrett, M. (2004). Representing the meaning of object and action
words: The featural and unitary semantic space (FUSS) hypothesis. Cognitive Psychology, 48, 422–
488.
Vinson, D., & Vigliocco, G. (2008). Semantic feature production norms for a large set of objects and events.
Behavior Research Methods, 40(1), 183–190.
Watzl, S. (2022). The ethics of attention: An argument and a framework. In S. Archer (Ed.), Salience: A
philosophical inquiry (pp. 89–112). Routledge.
Weinberg, J., Nichols, S., & Stich, S. (2001). Normativity and epistemic intuitions. Philosophical Topics,
29(1 & 2), 429–460.
Whiteley, E. (2022). Harmful salience perspectives. In S. Archer (Ed.), Salience: A philosophical inquiry
(pp. 193–212). Routledge.
Wiegmann, A., & Willemsen, P. (2017). How the truth can make a great lie: An empirical investigation of
the folk concept of lying by falsely implicating. Cognitive Science.
Willemsen, P., & Reuter, K. (2021). Separating the evaluative from the descriptive: An empirical study of
thick concepts. Thought: A Journal of Philosophy, 10(2), 135–146.
Wu, W. (2011). What is conscious attention? Philosophy and Phenomenological Research, 82(1), 93–120.
Wu, L., & Barsalou, L. (2009). Perceptual simulation in conceptual combination: Evidence from property
generation. Acta Psychologica, 132(2), 173–189.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
123
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center
GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers
and authorised users (“Users”), for small-scale personal, non-commercial use provided that all
copyright, trade and service marks and other proprietary notices are maintained. By accessing,
sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of
use (“Terms”). For these purposes, Springer Nature considers academic use (by researchers and
students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and
conditions, a relevant site licence or a personal subscription. These Terms will prevail over any
conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription (to
the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of
the Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may
also use these personal data internally within ResearchGate and Springer Nature and as agreed share
it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not otherwise
disclose your personal data outside the ResearchGate or the Springer Nature group of companies
unless we have your permission as detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial
use, it is important to note that Users may not:
use such content for the purpose of providing other users with access on a regular or large scale
basis or as a means to circumvent access control;
use such content where to do so would be considered a criminal or statutory offence in any
jurisdiction, or gives rise to civil liability, or is otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association
unless explicitly agreed to by Springer Nature in writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a
systematic database of Springer Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a
product or service that creates revenue, royalties, rent or income from our content or its inclusion as
part of a paid for service or for other commercial gain. Springer Nature journal content cannot be
used for inter-library loans and librarians may not upload Springer Nature journal content on a large
scale into their, or any other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not
obligated to publish any information or content on this website and may remove it or features or
functionality at our sole discretion, at any time with or without notice. Springer Nature may revoke
this licence to you at any time and remove access to any copies of the Springer Nature journal content
which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or
guarantees to Users, either express or implied with respect to the Springer nature journal content and
all parties disclaim and waive any implied warranties or warranties imposed by law, including
merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published
by Springer Nature that may be licensed from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a
regular basis or in any other manner not expressly permitted by these Terms, please contact Springer
Nature at
onlineservice@springernature.com
Article
Full-text available
We sometimes say about certain things (such as philosophical questions) that they are ‘deep’ or ‘profound’. But what does it mean exactly? Surprisingly, philosophers have been quite silent on this topic and there is neither consensual nor prominent conceptual analysis of ‘depth’. After surveying different proposals that have tried to capture depth in terms of explanatory depth, subject matter, emotional impact, understanding or complexity, we present the results of three studies in which we combine qualitative and quantitative methods to investigate the everyday concept of ‘depth’. Our results suggest that depth and profundity are generally characterized as (i) something with a meaning, (ii) that stimulates cognitive activity, to (iii) promote epistemic progress about (iv) important issues, thus (v) having an emotional impact. Based on these findings, we propose a first characterization of the ordinary concepts of depth and profundity.
Chapter
Full-text available
Schizophrenia and psychosis are associated with wide-ranging phenomenological disturbances. These, it is often said, involve ‘aberrant salience’. Others have sought to account for the relevant phenomenology in terms of ‘affordance’. In this chapter, we identify a shortcoming that is common to both approaches. There are many distinctions to be drawn between different kinds and different aspects of salience, some or all of which might prove clinically relevant. Consequently, terms such as ‘salience dysregulation’ and ‘aberrant salience’ lack the required discriminatory power. They should serve only as a starting point for the task of understanding anomalous experience in psychiatric illness and relating it to neuroscience. We go on to argue that recent appeals to experiences of ‘affordance’ fall short in the same way. Better, we suggest, to acknowledge the many subtly but importantly different ways in which human experience is permeated by a sense of the possible than to mask this complexity and diversity by settling for concepts that are insufficiently discerning.
Article
Full-text available
In much of the current academic and public discussion, conspiracy theories are portrayed as a negative phenomenon, linked to misinformation, mistrust in experts and institutions, and political propaganda. Rather surprisingly, however, philosophers working on this topic have been reluctant to incorporate a negatively evaluative aspect when either analyzing or engineering the concept conspiracy theory. In this paper, we present empirical data on the nature of the concept conspiracy theory from five studies designed to test the existence, prevalence and exact form of an evaluative dimension to the ordinary concept conspiracy theory. These results reveal that, while there is a descriptive concept of conspiracy theory, the predominant use of conspiracy theory is deeply evaluative, encoding information about epistemic deficiency and often also derogatory and disparaging information. On the basis of these results, we present a new strategy for engineering conspiracy theory to promote theoretical investigations and institutional discussions of this phenomenon. We argue for engineering conspiracy theory to encode an epistemic evaluation, and to introduce a descriptive expression—such as ‘conspiratorial explanation’—to refer to the purely descriptive concept conspiracy theory.
Article
Full-text available
Conceptual engineering is the method for assessing and improving our concepts. However, little has been written about how best to conceive of concepts for the purposes of conceptual engineering. In this paper, I aim to fill this foundational gap, proceeding in three main steps: First, I propose a methodological framework for evaluating the conduciveness of a given concept of concept for conceptual engineering. Then, I develop a typology that contrasts two competing concepts of concept that can be used in conceptual engineering—namely, the philosophical and psychological ones. Finally, I evaluate these two concepts of concept using the proposed methodological framework and I show that, when it comes to making conceptual engineering an actionable method, the psychological concept of concept outclasses its philosophical counterpart on all counts. This provides a baseline from which the concept of concept can be further improved for the purposes of conceptual engineering.
Chapter
Corpus analysis allows researchers to inform, illuminate, and investigate many problems. This chapter provides easy access to some of the central tools commonly used in corpus linguistics. After a short exploration of pre-built corpora and a brief literature review surveying corpus-analytic studies in philosophy, we illustrate these tools by running several corpus analyses on the term “conspiracy theory.” These analyses show that “conspiracy theory” is a strongly evaluative term. The reader of this chapter can follow each of the steps of the corpus analyses using the online material that is freely available.
Article
Beyond the context of cross-cultural pragmatics, it has been claimed that race and Rasse are not equivalent terms. The current paper seeks to establish the first known program of study to shed light on the extent to which race and Rasse differ. Corpora of US and German newspapers are used in this paper to sample mainstream race talk in the respective countries, and top collocates are analyzed along with a selection of examples. What is seen is that, while the historical context of slavery and the civil rights movement in the US and Nazism in Germany do seem to be reflected in the respective corpora, it is not so clear that the underlying meaning of race and Rasse are as different as some have claimed
Article
This paper details the content and structure of the folk concept of life, and discusses its relevance for scientific research on life. In four empirical studies, we investigate which features of life are considered salient, universal, central, and necessary. Functionings, such as nutrition and reproduction, but not material composition, turn out to be salient features commonly associated with living beings (Study 1). By contrast, being made of cells is considered a universal feature of living species (Study 2), a central aspect of life (Study 3), and our best candidate for being necessary for life (Study 4). These results are best explained by the hypothesis that people take life to be a natural kind subject to scientific scrutiny.