Content uploaded by Michael Hemmingsen
Author content
All content in this area was uploaded by Michael Hemmingsen on Mar 06, 2024
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rjps20
Journal of the Philosophy of Sport
ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/rjps20
Mapping the terrain of sport: a core-periphery
model
Michael Hemmingsen
To cite this article: Michael Hemmingsen (06 Mar 2024): Mapping the terrain of sport: a core-
periphery model, Journal of the Philosophy of Sport, DOI: 10.1080/00948705.2024.2317876
To link to this article: https://doi.org/10.1080/00948705.2024.2317876
Published online: 06 Mar 2024.
Submit your article to this journal
View related articles
View Crossmark data
Mapping the terrain of sport: a core-periphery model
Michael Hemmingsen
International College, Tunghai University, Taiwan, Republic of China
ABSTRACT
In this paper, I propose a new way of dening sport that I call a ‘core-periphery’
model. According to a core-periphery model, sport comes in degrees – what
I refer to as ‘sport-likeness’ – and the aim of the philosopher of sport is to chart
those dimensions along which an activity can be more or less a sport. By
introducing the concept of sport-likeness, the core-periphery model compli-
cates the picture of what is or is not a sport and encourages philosophers
interested in dening sport to engage with the social sciences in exploring the
extension of the term sport in common usage. In this paper I present the results
of a small survey about attitudes to sport, and use it to illustrate how a core-
periphery denition of sport would proceed.
KEYWORDS Defining sport; movement compression; constitutive rules; facilitative rules; core-periphery
Introduction
In this paper, I propose a new way of dening sport – distinct from existing
approaches such as essentialism, Wittgensteinian family resemblance or focal
meaning – that I call a ‘core-periphery’ model of sport.
1
According to a core-
periphery model, sport comes in degrees – what I refer to as ‘sport-likeness’ –
and the aim of the philosopher of sport is to chart those dimensions along
which an activity can be more or less a sport. In this view, there are central –
core – activities that are indisputably sports – that possess all of the essential
characteristics of sport to a sucient degree – and then there are activities
that lack one or more of the core characteristics of sport, or possess them to
a lesser degree than ‘core’ sports, and which are therefore less sport-like.
By introducing the concept of sport-likeness, the core-periphery model
complicates the picture of what is or is not a sport and, like a family resem-
blance model, encourages philosophers of sports to ‘look and see’ how the
term is used in practice. The core periphery model therefore encourages
philosophers interested in dening sport to engage with experimental meth-
ods to discover the extension of the term sport in common usage. From there,
CONTACT Michael Hemmingsen mhemmingsen@thu.edu.tw International College, Tunghai
University, P.O. Box 898, Taiwan, Taichung 40704, Republic of China
JOURNAL OF THE PHILOSOPHY OF SPORT
https://doi.org/10.1080/00948705.2024.2317876
© 2024 IAPS
philosophers of sport can then attempt to determine those features pos-
sessed by the core of sport.
In this paper I start by outlining the core-periphery model of sport and
distinguishing it from other approaches. Then, I present the results of a small
survey and use it to illustrate how a core-periphery denition of sport would
proceed. This survey is too limited in scope for the analysis to act as
a denitive statement about the nature of sport. Instead, I use the survey to
illustrate how the dening of sport ought to be undertaken using the core-
periphery approach. In doing this, I tentatively draw out ve characteristics of
core sports: (core) sports 1) are essentially physical games; 2) are structured
by constitutive rather than facilitative rules; 3) have the central aim of testing
skills; 4) are played competitively; and 5) are open to all.
However, though this survey draws on everyday understandings of the
term ‘sport’, philosophers of sport are not unthinkingly beholden to common
usage even when using a core-periphery approach. There are therefore
certain issues that remain that require traditional philosophical methodolo-
gies to settle. For instance, it is an open question whether a good denition of
sport should include only internal characteristics, or whether it is appropriate
to include external characteristics. Working out whether we ought to include
external features or only internal ones is a matter for philosophers, rather
than social scientists, and I touch on this issue briey at the end of the paper.
Sport-likeness and core-periphery
In philosophical discussion about sport, borderline cases abound. There are
disputes, for instance, about the status of ‘mind sports’ such as chess and
bridge (Kobiela 2018); ‘nature sports’ such as hiking (Howe 2008, 2018, 2019;
Krein 2014, 2015); activities with a central role for animals, such as equestrian
(Holt 2023); activities involving machines, such as motor sports or shooting
(Llorens 2017; Parry 2019); ‘bar sports’ such as darts or billiards (Paddick 1975);
eSports (Hallmann and Giel 2017; Hemmingsen 2021; Hemphill 2005; Holt
2016; Jenny et al. 2016; Kane and Spradley 2017; Llorens 2017; Parry 2019;
Witkowski 2012); martial arts (Allen 2013); judged sports such as synchronised
swimming, gymnastics, gure skating (Hurka 2015); and even athletics
(Berman 2013).
A question for philosophers of sport is how we account for these
various activities. However, the answer to the question of what is or is
not a sport depends crucially on how we approach dening terms like
‘sport’ in the rst place. In this paper, I propose a core-periphery
approach. In a core-periphery approach, ‘sports’ possess certain fea-
tures. However, while there is a subset of activities – the core – that
enjoy all of these features to a required degree, there are other activ-
ities that possess only some of them, or have certain of these
2M. HEMMINGSEN
characteristics to a reduced extent. These activities are still sports, but
are located on the ‘periphery’ of the denition: that is, they are sports,
but less so than the core.
As a result of taking this approach, instead of asking whether or not an
activity is a sport, a core-periphery approach instead asks us to consider
how ‘sport-like’ an activity is (and why). For instance, there is a dierence
between darts and cricket that can give rise to an reluctance to put them
on the same footing. But it doesn’t seem plausible to eliminate darts from
the sports family entirely. With the concept of ‘sport-likeness’ we can see
the distinction between darts and cricket as a matter of degree, rather
than being an either/or question. In this case we would say something like
‘darts is a sport, it’s simply less of a sport than cricket’ and then – with
reference to a robust core-periphery model of sport – be able to explain
precisely why.
In addition, some of the denitional criteria of sport themselves
admit of no obvious sharp dividing line. For example, essential physical
movement is a frequent candidate for a dening feature of sport. But it
seems obvious that activities can have more or less essential physical
movement, and determining the point at which an activity switches
between sport and non-sport on this basis is not easy. However, if we
understand ‘sport’ as a gradated category, rather than as all-or-nothing,
we can avoid drawing an arbitrary line. Thinking in terms of sport-
likeness allows us to describe an activity with more essential movement
as more sport-like than one with less, and then explain precisely why it
is so in detail. There is no further question – is this (or is this not)
a sport? – that needs to be answered.
An analogy to this way of thinking about sport is Derek Part’s
approach to personal identity (Part 1986). For Part, what we care
about is not whether a future being is or is not us in a binary sense, but
rather how similar the future being is to us.
2
For Part, there is simply no
interesting answer to the question ‘is that future being me?’ when asked in
a binary way. In fact, by focusing on that question, we end up eliding the
question of similarity, which ought to be our concern. Similarly, if we
understand sport through the lens of ‘is this activity a sport?’ we make
the same kind of mistake. But if we understand activities as a collection of
similarities and dierences – which is the focus of a core-periphery
account interested in sport-likeness – we can concentrate on charting
the specic qualities of activities broadly within the wider sport family in
detail, rather than giving undue attention to whether an activity is or is
not a sport in a nal, conclusive sense. A core-periphery account of sport
therefore emphasises describing the terrain of sports as clearly as possible,
rather than focusing on answering the question of which activities are ‘in’
and which are ‘out’.
JOURNAL OF THE PHILOSOPHY OF SPORT 3
Competing approaches
In this section, I contrast the core-periphery approach to denition with other
dominant approaches, with the aim of a) clarifying the core-periphery
approach; and b) foregrounding some of the advantages of the core-
periphery approach as compared to other approaches.
Essentialism
For the essentialist, there are certain necessary and sucient characteristics
that an activity must possess to be a sport. If an activity lacks even one, it can
no longer be considered a sport. However, this way of dening sport creates
two problems.
First, common usage of the term ‘sport’ makes it extremely dicult to pin
down essential characteristics. While I certainly don’t think that an activity is
a sport merely because people commonly refer to it using this word – it’s
important, I think, to recognise that the focus is not on the word but on the
concepts that stand behind it, and it’s the philosopher’s job to clarify those
concepts even when it entails rejecting aspects of common usage – the
essentialist approach risks becoming overly stipulative. That is, it can be
insuciently attentive to how words are used and fail to meaningfully reect
the kinds of concerns that motivate us to attempt to delimit and describe
concepts like sport in the rst place.
A core-periphery model, by contrast, lets us lay out more carefully the
important dimensions or characteristics of sport (the core) without insisting
that activities that fall short of that core need to be cut out of the picture
entirely. This allows us to focus more on describing important dierences and
commonalities between activities commonly thought of as sports, while
deemphasising overeager stipulation.
Second, and relatedly, the essentialist approach has no room for matters of
degree. However, when it comes to certain commonplace (and common
sense) aspects of sport, such as the necessity of physical movement, it is
simply not obvious that sharp dividing lines can be drawn non-arbitrarily.
Family resemblance
In certain respects, the core-periphery model is more similar to
a Wittgensteinian family resemblance approach than to essentialism.
According to the family resemblance view, there are no essential character-
istics possessed by a concept like sport. Rather, the emphasis becomes about
identifying ‘overlapping . . . characteristics that form a complex network of
similarities’, a ‘“family of resemblances” in which the degree of similarity is
4M. HEMMINGSEN
closest when we consider adjacent members of the family and furthest apart
when we consider distant members of the family’ (Morgan 1977, 17).
However, in a family resemblance model it is possible to ‘chain’ features:
for instance, because A possesses features (i), (ii) and (iii), B possesses (iii), (iv)
and (v), and C possess (iv), (v) and (vi), then even though there is no single
feature in common between all three, we are still able to understand them as
instances of a certain category (see Figure 1).
In a core-periphery model, by contrast, there is a core that possesses
certain features – say, (i), (ii), (iii), (iv) and (v) – and then a range of activities
that count as instances of that kind (though to a lesser extent) by virtue of
possessing one or more of those features. Furthermore, the features can be
possessed by activities to various degrees, meaning that individual activities
that share the same features can be considered closer or further from the
‘core’ denition (see Figure 2).
Figure 1. Family resemblance.
Figure 2. Core-periphery.
JOURNAL OF THE PHILOSOPHY OF SPORT 5
The advantage of the core-periphery approach is that it captures better the
intuition that some activities are more obviously or centrally sports than
others. The family resemblance view has diculty with this since, unlike the
essentialist view, it tends to resist the idea that certain criteria are more at the
heart of a denition than others.
Furthermore, a core-periphery approach does not chain features,
a practice that can end up capturing too much. As Johan Steenbergen points
out, an uncritical use of the family resemblance model can lead to relativism,
where ‘sport can not be demarcated from non-sport because there are no
limiting criteria’ (Steenbergen 2001, 40).
Hence, essentialism, in William J. Morgan’s words, “invariably is either
too broad (such that activities not commonly thought of as sportive
enterprises are granted admission into the realm of sport) or too narrow
(such that activities commonly regarded as sportive ones are denied
admission into the realm of sport)” (Morgan 1977, 16). However, a family
resemblance view, while allowing more ambiguity than an essentialist one,
arguably makes concepts like ‘sport’ too ambiguous and imprecise. A core-
periphery model, by contrast, has the best of both worlds: like the essen-
tialist, it can point to the specic features that demarcate sport from non-
sport, while adopting the family resemblance view’s move away from
necessary and sucient criteria.
Focal meaning
For a focal meaning view, while terms are used in many dierent ways, each
of those ways ties back in some sense to a ‘primary meaning’ that ‘serves as
the normative guide for systematically ordering all its other meanings’
(Morgan 1977, 26). Morgan uses the term ‘medicine’ to illustrate this idea:
The concept medicine . . . can be used, for example, to refer to the science of
medicine, or to a medical man, or to a medical instrument . . . For all these
respective uses of the concept refer back to one common core of meaning which
is to be regarded as the primary meaning of the term, in this case the meaning
indicated by the science of medicine itself. That is to say, we call an individual
a medical man only because he is a practitioner of the science of medicine.
Similarly, an instrument is designated as a medical instrument only in virtue of
the fact that it is used by a medical man to further implement the science of
medicine. In each case then, we are entitled to predicate the concept of medicine of
some person or thing because it refers back to, and variously qualies, the primary
meaning denoted by the science of medicine. (Morgan 1977, 21–2)
Just as the science of medicine is taken here to be the primary meaning of
‘medicine’, for a focal meaning approach to sport, something would act as the
focal meaning of the term ‘sport’, with other things related to that focal
meaning referentially.
6M. HEMMINGSEN
However, a core-periphery model diers from the focal meaning view in
that it is more interested in criteria than in reference. For instance, in the
medicine example we go from the science of medicine to things – medicine
man, medical instrument – that refer back to the core concept of medical
science. Focal meaning is about extending a core meaning into other
domains (practitioners, instruments). However, a core is about nding the
centrally important features within a single domain (activities). In the core-
periphery model, then, the issue is whether activities in the same domain
possess certain features (and to what degree), rather than how a primary
meaning is extended into other domains.
In certain respects, however, a core-periphery model takes something
from all three of these approaches: the core, in a sense, is an adoption of
the idea of focus from focal meaning; the comfort with ambiguity and rejec-
tion of necessary and sucient conditions is taken from the family resem-
blance view to allow for the existence of a periphery; and it borrows an
interest in delimitation from the essentialist view to set the boundaries of
that periphery (in the sense that at a certain point activities will be so lacking
in core features that they are no longer examples of sport at all).
Core-periphery and social science
What features of an activity are those that matter in terms of its sport-likeness ?
According to a core-periphery model, some features are likely to be absolutely
required, and some might be optional. Of the optional criteria, it’s an open
question the extent to which they matter (for instance, precisely how much less
of a sport is an activity that lacks one or more of the optional criteria?).
Furthermore, many of the criteria are likely to be a matter of degree, and it is an
open question precisely how sharp the drop-o the sport-likeness of an activity is,
as the feature reduces.
It’s not obvious that these problems can be resolved ‘from the armchair’. The
question of dening sport from a core-periphery perspective therefore ends up
being at least partly empirical, a matter of how the word is used in practice.
3
The rest of this paper is therefore focused on demonstrating how a core-
periphery approach might be applied, combining the collection of empirical data
about how people commonly understand the term ‘sport’ with a philosophical
analysis from a core-periphery perspective. I do this by presenting and analysing
the results of a survey conducted in 2022 on attitudes towards various sports-like
activities.
I want to emphasise that the remainder of this paper is intended as an
illustration of the core-periphery model, rather than as presenting any nal
conclusions about how sport should be dened from a core-periphery perspec-
tive. I hope to conduct more research in this vein in the future on a larger scale,
ideally in collaboration with experimental philosophers. Here, however, my
JOURNAL OF THE PHILOSOPHY OF SPORT 7
central goal is to demonstrate how we might approach dening sport from
a core-periphery perspective in general terms.
Survey methodology
My survey received 215 responses in total. The respondents were predomi-
nantly university students, either students in my own International College at
Tunghai University in Taiwan, or the students of colleagues at universities in
the USA, Canada, China and South Korea, with a cluster of responses from the
colleagues of a friend at a video game design company in New Zealand.
As a result of the sampling method the respondents skewed younger than
the population average: most (73%) respondents were between 18–24 years
old. Slightly under half (49%) were from English-speaking countries (the US
[11%], Canada [25%], the UK [3%], New Zealand and Australia [8%], South
Africa [1.5%]); the remaining respondents were predominantly either from
East Asia (~21% in total, including Taiwan [16%], China [2%], Japan [1%] and
Korea [1.5%]) or South East Asia (~20% in total, including Indonesia [12%],
Thailand [3%], the Philippines [2.5%], Cambodia [1.5%], Vietnam [1%] and
Singapore [0.5%]). The remaining respondents were from South Asia (2.5%),
Eastern Europe (2.5%), Africa (2%), the Middle East (2%) and the EU (2.5%).
The survey was therefore dominated by English-speaking and Asian respon-
dents, with non-native-English speaking or non-Asian respondents making
up only around 7% of the total.
The questionnaire was conducted online, using the website SurveyMonkey.
The survey was described to respondents as ‘part of a project to how people
understand the concept of “sport”’. It contained four questions:
(1) An open question about the respondent’s nationality;
(2) A question about the respondent’s age, with options being ‘Under 12
years old’; ’12-17 years old’; ’18-24 years old’; ’25-34 years old’; ’35-44 years
old’; ’45-54 years old’; ’55-64 years old’; ’65-74 years old’; and ’75 years or
older’;
(3) ‘Assuming that we are talking about competitive versions of the activ-
ities listed below, please evaluate your level of condence that they
should be classied as “sports”’. Answers to this question were given
on a 7-point Likert Scale; the activities were denitely (3), probably (2),
may be (1), may not be (−1), are probably not (−2), or are denitely not
(−3) a sport, with a neutral option (0) in the middle (‘I am not sure
whether or not this is a sport’).
(4) ‘What is your level of interest in the following activities?’ Answers to
this question were given on a 5-point Likert scale: not at all interested
(1); not very interested (2); a little interested (3); somewhat interested (4);
very interested (5).
8M. HEMMINGSEN
The activities listed in questions (3) and (4) were Soccer; Rugby; Cricket;
Tennis; Badminton; Baseball; Golf; Ice Hockey; Basketball; Table Tennis;
Synchronised Swimming; Rowing; Toboganning; Fencing; Piano; Formula
One Racing; Gymnastics; Ballroom Dancing; Diving; Yachting; Karate; Judo;
Figure Skating; League of Legends; Counter-Strike: Global Oensive; Fornite;
Swimming; Running; Shooting; Archery; Horse Riding; Darts; Weightlifting;
High Jumping; Javelin; Shot Put; Chess; Cycling; Fishing; Billiards; Pick Up
Basketball; and Street Football. The order of the options was randomised for
each respondent.
In choosing these activities, I attempted to capture as wide a range of
sports or potential sports as possible, especially with mind to cases where
there is active debate about whether a general class of activity should be
included as sport in the rst place, e.g. eSports, mind sports, bar sports,
animal sports, judged sports, and so on. The notable class missing in the
survey is nature sports.
One acknowledged problem with this survey design is with question
(3). In this question, the neutral option is ambiguous: it could mean ‘I
don’t know enough about this activity to rate it’ or ‘I know about the
activity but am unsure how to categorise it’. These two potential
responses should have been presented as separate options. In the end,
due to this ambiguity I chose to exclude all ‘0’ answers from the results.
In a few cases, this did have some eect on the scores – cricket goes
from an average of 2.52 with the ‘0’ responses included to 2.72 with
them excluded, and tobogganing goes from 0.92 to 1.17 – but in general
including or excluding the ‘0’ responses had minimal impact on the
results. In the cases where it does have an eect, the activities tend to
be those that are little known outside certain countries and so it made
more sense to treat such responses as ‘I don’t know enough about this
activity to rate it’ rather than ‘I don’t know how to categorise this
activity’.
Another design problem is that questions (3) and (4) used an ordinal Likert
scale rather than an interval scale, which would have allowed me to properly
identify dierences along intervals. While I do not think that this undermines
the goal of the paper – e.g. illustrating in general terms how a core-periphery
model would work – it is certainly an element of the survey design that would
need to be corrected for any future research.
Survey results
Of all 42 activities listed, 29 received an average response of ‘2’ or above
(between ‘is probably a sport’ and ‘is denitely a sport’). However, there was
signicant diversity within this range. Hence, I divided the ‘2’ range into
JOURNAL OF THE PHILOSOPHY OF SPORT 9
several groups: core activities (2.78 and above), slightly controversial activities
(2.5 to 2.77), and somewhat controversial activities (2.25–2.49).
The core activities comprised (in descending order): Soccer, Badminton,
Tennis, Basketball, Baseball, Rugby, Ice Hockey, Swimming, Running and
Cycling. The second group contained Cricket, Gymnastics, High Jump,
Rowing, Figure Skating, Table Tennis, Judo, Fencing, Karate, and Diving.
Finally, we have Golf, Javelin, Weightlifting, Archery, Shot Put, Street
Football, Synchronised Swimming, Pick Up Basketball and Horse Riding as
somewhat controversial activities (see Figure 3).
After this, there is a large and quite noticeable gap between lowest score
in the ‘2’ range (2.25) and the next highest score, found in the ‘1’ range
(between ‘may be a sport’ and ‘is probably a sport’), starting with formula one
racing at 1.49, followed closely by shooting (1.44), with another small jump to
billiards (1.27), then tobogganing, darts and ballroom dancing following far
behind at 1.17, 1.09 and 1.07 respectively. Yachting follows closely behind
this 0.95.
Around 0 we nd a small constellation ranging from 0.03 to 0.26, consist-
ing of shing, chess and the eSport, Counter-Strike. This is followed by a group
of two other eSports – Fortnite (−0.63) and League of Legends (−0.51) – trailed
far behind by competitive piano playing (−1.80) (see Figure 4).
The core activities − 2.78 and above – also display a very low standard
deviation, ranging from 0.51 to 0.76, with an average of 0.65. As the mean
Figure 3. Graph of averages (2+ Range).
10 M. HEMMINGSEN
score reduces in the slightly controversial and somewhat controversial activ-
ities, the standard deviation increases as well, from 0.84 at the higher end of
the slightly controversial activities, to 1.32 at the lower end of the somewhat
controversial activities. This standard deviation increases further as the score
goes down (see Figure 5).
This is what a core-periphery model would expect. There are likely to be
several core features of sport, and these features can come apart. When it
comes to peripheral activities, dierent features will be pulling in dierent
directions. If respondents are paying attention to, or emphasising, dierent
features of the core, then they will also dier on whether the presence of
absence of a certain feature is sucient to rule that activity in or out, and this
variety of responses will increase the standard deviation.
The exception to the general trend for increased standard deviation as the
scores drop is piano. Given that piano is also the stand-out lowest score (−1.8,
with the next lowest being −0.63), this makes sense: piano is so far from the
core – it lacks so few of the core features of sport – that the uncertainty about
which features to emphasise no longer arises.
Analysis
The task for a core-periphery model of sport is to draw from the data the
features that are present in the activities in the core of sport, but that are
missing or attenuated in activities that are considered to be more peripheral.
This is the empirical dimension of the core-periphery approach. We also need
to ‘tidy’ the results. This may involve correcting local confusions, such as
Figure 4. Graph of averages (All).
JOURNAL OF THE PHILOSOPHY OF SPORT 11
where activities are rated higher or lower than can be explained without
resorting to idiosyncratic causes. Cricket is an example of this: the fact that it
is almost unknown outside the Commonwealth better explains its low score
compared to similar-seeming activities such as baseball than assuming that it
diers in some signicant, fundamental way.
But tidying the results may also involve making a philosophical case for
a certain view about what kinds of features are appropriate for including in
a denition in the rst place. I do this below in my discussion of whether or
not to include two features that do seem to genuinely explain widespread
dierences in scores across groups of activities: leisure and access. Whether or
not certain criteria are the kinds of criteria that should ground a denition of
sport is not able to be settled by the analysis of empirical data: it is
a philosophical question, not a social scientic one.
For the remainder of this paper I attempt to abductively draw out the
principled features that best explain the survey results in terms of a core of
activities that possess all features of sport and a periphery that lacks or
attenuates one or more of those features; I do my best to account for outliers
that seem to be contingent rather than principled; and I identify the features
for which there is uncertainty about whether they ought to be included in
a consistent, coherent core-periphery denition of sport, even if they seem to
be part of a common sense understanding.
4
The analysis of the results involves identifying which features are pos-
sessed by ‘core’ sports, but that are absent or reduced in sports that have
Figure 5. Standard deviation vs. score.
12 M. HEMMINGSEN
received a lower rating. It seems to me that there are several criteria that core
sports possess (and that other activities do not) that I think help to explain the
results:
(1) Core sports are essentially physical.
(2) Core sports are structured by constitutive rather than facilitative rules.
(3) The competition in core sports has as a central aim the testing of skills.
(4) Core sports are played competitively, and not merely for leisure.
(5) Core sports are open to all.
I will discuss each of these in turn.
Physical games and essential physicality
Bernard Suits claims that sports are games, and of a particular kind: they are
‘games of physical skill’ (2005, 2). Sports are those games in which physical
skill is essential – rather than incidental – to the activity itself. It cannot merely
be that the participants perform physical activities in the game: monopoly
players make physical movements when they roll the dice or move their piece
around the board, but monopoly is not a sport. Rather, in sports ‘the outcome
is dependent, to a certain degree at least, upon the physical prowess exhib-
ited by the participants’ (Meier 1988, 13). In Monopoly greater or lesser skill in
moving pieces around the board does not contribute to the outcome of the
game: these movements are therefore ‘incidental, peripheral, and of little or
no consequence’ (Meier 1981, 83). By contrast, the outcome in the game of
baseball is ‘necessarily and signicantly determined by the demonstration
and exercising of physical skill or prowess’ (Meier 1981, 85).
However, while essential physicality has been a central idea in discussions
of sport (Paddick 1975; Schieman 2016; Parry 2019), precisely what is meant
by ‘essential physicality’ is still somewhat unclear. There are two plausible
accounts of essential physicality that seem to account for the survey results:
the gross vs. ne movement distinction (Loy 1968) and movement compres-
sion (Hemmingsen 2023a).
Gross vs. ne movement focuses on the quality of the movement itself:
whether it involves large movements utilising major muscle groups, or
whether it is about exact movements that only use small muscles, usually in
the hands and ngers. This distinction has been criticised by some as ‘arbi-
trary and counterproductive’ (Meier 1981, 84), since it can be dicult to draw
a sharp boundary between activities that utilise ne or gross motor skills.
However, while drawing a sharp distinction may matter to an essentialist
denition of sport – since activities have to be either ruled in or ruled out
denitively – a core-periphery approach is comfortable with the idea that the
gross/ne distinction is a spectrum along which sports activities can lie.
JOURNAL OF THE PHILOSOPHY OF SPORT 13
The survey results do provide some support for the gross vs. ne
movement way of understanding essential physicality: despite being
highly physical at the professional level, eSports favour ne over gross
motor control, and this is reected in the extremely low scores of the
eSports included in the survey: Counter-Strike (0.02), League of Legends
(−0.51) and Fortnite (−0.62). This distinction may also account for the
comparatively low scores of horse riding (2.24) and formula one (1.48)
(though I think there may be confounding factors here as well). Since
horse riding requires more gross physicality (in the sense of gross phy-
sical control, if not physical movement) than formula one, but not as
much as, for example, soccer or basketball, then it makes sense that
horse riding would be higher than formula one but lower than a core
sport like soccer.
Finally, archery (2.46) had a much higher score than shooting (1.44).
Though both involve whole body control, the drawing of the bow in archery
adds a degree of gross movement that is not present in shooting, thereby
potentially explaining archery’s higher score.
By contrast, movement compression is about the eect of the movement
on the game, e.g. the degree to which ‘the quality of the initial action
translates to a dierence in outcome in the activity along various dimensions’
(Hemmingsen 2023a, 7). For instance, in cricket,
extremely small dierences in where I strike the ball with my bat, the angle of
the bat, the speed of movement, and so on, aect where that ball ends up
going. When I play a stroke, there’s no aspect of my movement that fails to be
relevant to what subsequently happens. (Hemmingsen 2023a, 7)
By contrast, in the eSport Counter-Strike,
clicking the left mouse button res your weapon. Many details of this action
matter to the outcome, such as the precise timing of the click, and where the
cursor is pointing, etc. However, it doesn’t matter at all how you click the
button; it doesn’t matter if you press it softly or with force, [or] what nger
you use . . . the quality of your . . . movement here is compressed into a single
outcome. (Hemmingsen 2023a, 8)
Movement compression seems to be consistent with the survey results as
well. eSports involve signicantly more movement compression than any
other activity from the survey, thereby explaining the extremely low scores
of Counter-Strike, League of Legends and Fortnite. Similarly, the fact that
a person’s actions are mediated in horse riding and formula one (through
the horse and car respectively) introduces a level of movement compression
(though less than in eSports). As for the dierence in the scores of archery and
shooting,
14 M. HEMMINGSEN
in archery, the ight of the arrow depends not only on how the bow is held and
where it’s pointed, but also crucially on how the bow is drawn. In shooting, while
it clearly matters in certain ways how one pulls the trigger – shooters need to pull
the trigger in such a way as to ensure that the gun doesn’t move inappropriately,
for instance – nonetheless the trigger-pull is a point of movement compression. . .
The ight of the bullet occurs identically regardless of whether the trigger is
pulled gently or with force, in just the same way as a mouse click in League of
Legends or Counter Strike has the exact same eect regardless of the quality of the
movement that goes into it. (Hemmingsen 2023a, 15)
Hence, while it seems likely that essential physicality makes a dierence to an
activity’s sport-likeness, it isn’t clear precisely how to cash out this concept. Of
the two, the gross/ne distinction is more concerned with the athleticism of
an activity. There can be highly athletic activities that also involve high levels
of movement compression: a video game such as Dance Dance Revolution, for
instance. But typically, more gross movement will correspond to greater
athleticism. So, if essential physicality is connected in some way to athleti-
cism, then this would incline us towards the gross/ne distinction as the best
way of cashing this idea out.
However, especially when considering the constitutive/facilitative distinc-
tion discussed below, it is not obvious that this is the way to go. Further
research needs to be undertaken to determine whether the gross/ne dis-
tinction or the concept of movement compression is a better account of
essential physicality (or indeed whether it is both, or neither).
Constitutive vs. facilitative goals
Athletics are often considered obvious instances of sport. However, this is not
wholly uncontroversial, and there has been some discussion in the philoso-
phy of sport about how to classify athletics. Mitchell N. Berman, for example,
states that he
[does] not nd it jarring to conclude that there is a potentially signicant
distinction between, say, baseball, soccer, and tennis on the one hand and
the 100-meter dash, the high jump, and weightlifting on the other. (Berman
2013, 172)
Whereas the former are sports, he thinks, the latter are not. And it seems that
the results of the survey provide some support for Berman’s reluctance to
include athletics within the core of sport. Of the athletics included in the
survey, only swimming (2.87), running (2.82) and cycling (2.78) are found
within the top group. Even then, they are on the very bottom of the group,
and are lower than all traditional ball/ball-adjacent games, excepting cricket
(2.72),
5
table tennis (2.58), golf (2.49), street football (2.41) and pick up
basketball (2.31) (all of which are special cases that I will address later in the
paper).
JOURNAL OF THE PHILOSOPHY OF SPORT 15
Aside from swimming, running and cycling, which are at the lower end of
the top category, the other athletics in the survey include gymnastics (2.68),
high jump (2.65), rowing (2.63), javelin (2.47), weightlifting (2.47), and shot
put (2.41). These scores are quite notably lower than most of the ball/ball-
adjacent games included in the survey. It seems plausible, then, that athletics
lack some feature that puts such activities outside the core.
The sport/athletics distinction cannot be a matter of the level of skill, or
essential physicality, or athleticism of the activity. Some sports will involve
more skill, essential physicality or athleticism that some athletics (basketball
seems more athletic than the hammer throw, for instance); and conversely
some athletics will involve more skill, essential physicality and athleticism
than some sports (compare the average baseball player to a decathlete).
Rather, the basic issue with athletics is that the player’s movements are non-
conventionally eective. As Barry Allen puts it, ‘a runner works a track as
a medium; it receives the runner’s exertion and produces aimed-for eects.
These are real eects that do not depend on a convention’ (2013, 243). In
other words, when a runner learns to run fast, they’re learning movements
whose eectiveness does not depend in any way on the established conven-
tions of the activity of running. Running fast is running fast; and doing so is
eective outside of competitive running. As Kevin J. Krein puts it, ‘High jump,
and other sports like it, are designed in a way that allows us to measure
human athletes against each other. High jump is about being able to jump
higher than another human’ (Krein 2014, 198), nothing more, nothing less.
However, the movements in ball games are to a much greater extent
conventionally eective. A soccer play learns to pass the ball quickly and
accurately to other players. A baseball pitcher learns to throw a baseball
quickly and accurately. A tennis player learns to hit a tennis ball with
a racquet with enough topspin to keep the ball from landing beyond the
designated play area. None of these movements are eective outside the
games of which they’re a part. Players of these games may develop certain
real-world-applicable skills incidentally, but there is no real-world environ-
ment in which a tennis stroke, for example, is meaningful and eective in and
of itself. But runners are the best at running, cyclists the best at cycling,
swimmers the best at swimming, and so on. We don’t need a competitive
game to understand and appreciate these movements, even if we need
games to determine who is the best at them.
To put this issue in Berman’s terms, athletics are activities in which the
rules are facilitative rather than constitutive (Berman 2013, 166). We certainly
need rules in running competitions. Imagine, Berman says, what a running
race would look like without formal rules:
Questions regarding the conduct of the competition would naturally arise,
whether at the outset or over time. Precisely what distance should we run?
16 M. HEMMINGSEN
Should we run at the same time or seriatim? How many tries do we get? . . . How
do we know when to start? What if one of us starts too soon? Is that disqualify-
ing or does he get additional chances? If competitors are more or less tied at the
race’s conclusion, is the winner the one who rst touches the line or who rst
completely crosses it? (Berman 2013, 165–6)
We clearly need rules for athletic competition to be possible. However, the aim
of the rules is not to give rise to a game – to, as Berman puts it, ‘make realization
of the prelusory goals harder’ (Berman 2013, 166), as we nd in cricket or
soccer – but rather merely to most fairly facilitate the testing of these prelusory
goals. Hence, we can describe the rules as ‘facilitative’. By contrast, without a set
of rules, there is no such thing as soccer, or baseball, or tennis: these rules
constitute the very activity itself; hence ‘constitutive’ rules. I think ordinary
language use in English tracks this distinction: we play sports but do athletics.
This not only explains athletics’ lower scores in comparison to ball sports,
but it also explains the scores for the martial arts included in the survey:
fencing (2.56), judo (2.54), karate (2.51), and archery (2.46). As Allen puts it, in
martial arts ‘violence is essential to the art and its training’ (Allen 2013, 247).
Even when never used to harmful eect in the competitive activity itself,
to train [in a martial art] you have to understand how the techniques function
combatively. You have to know, performatively, where the energies go to make
a technique an irresistible response to violence . . . Everything about martial arts
movements is expressively purposive. It is not a game. No movements are
symbolic or merely graceful. There is an instrumentality being trained, and its
purpose is violence. (Allen 2013, 247)
When this fails to be true of a martial art, we might think of that activity as
a form of dance rather than as a martial art (Allen 2013, 247). In this respect,
the movements in martial arts are non-conventionally eective, and rather
than constituting the activity, the rules in competitions are instead to facil-
itate the prevention of injury, the showing of respect, and to enable the
purposive movements to be tested fairly.
In summary, though the constitutive vs. facilitative rule distinction does
not seem to be a weighty one when it comes to the sport-likeness of an
activity (there are many highly-rated activities that have facilitative rather
than constitutive rules), it arguably does make some dierence, and explains
the clear grouping of ball sports – all games – at the top, with athletics/
martial arts grouped slightly, though denitively, below this group. Cycling,
running and swimming are the highest of the athletic activities, but even
there they are rated lower than the ball sports, excepting those such as cricket
(which joins the other ball sports decisively when the score is restricted to
countries that are familiar with it), and the ‘leisure’ sports such as golf, table
tennis, street football and pick up basketball, which (I argue) are a special
category.
6
JOURNAL OF THE PHILOSOPHY OF SPORT 17
Tests of skills
Core sports are tests of skills. The skills being tested in sports are physical ones
(thanks to criterion [1]); however, essential physicality (1) is about whether
physical movement is integral to the game, whereas the testing of (physical)
skills (3) is about what is being evaluated. For instance, ice skating is essen-
tially physical: there is no version of ice skating that is not a physical activity;
what’s more, its physicality is essential rather than incidental. However, what
is being evaluated in ice skating is not merely the physical skills of skating,
there is also an aesthetic element. In this sense, ice skating might be thought
of as the production of aesthetic beauty via physical skill, rather than merely
a test of physical skill and nothing more.
Similarly, activities in which luck plays a signicant role – such as compe-
titive shing – are games in which skill is subordinated to random chance,
even if the activities remain essentially physical.
Additionally, activities in which the quality of the equipment matters to
a signicant degree – such as yachting, an essentially physical activity – are
arguably less of a test of skill alone; the fact that equipment diers between
competitors means that not only the competitor’s skill aects the outcome.
The idea that sports must centrally test skills therefore has several dimen-
sions. First, it asks us to consider the degree to which the competition is truly
between human beings. For instance, while the quality of the bicycle matters
to cyclists, and certain paraphernalia – such as swim caps – are a matter of
ongoing debate, generally speaking all competitors in cycling and swimming
start on an equal footing. But in activities such as formula one (1.48) and
yachting (0.95), the quality of the equipment can be a decisive factor. In
yachting, for instance, enormous sums are spent on designing and building
the yachts, with technological developments providing signicant advan-
tages. It is possible, then, that the decisively lower scores in these activities
is due to there being an unearned advantage during the active part of the
competition; the struggle is as much between the designers of the machines
as the humans engaged in the competition on the day, and that it therefore is
not exclusively a test of the skill of the participants.
Second, a higher role for chance undermines the activity as a test of skills.
This can perhaps be seen to some extent in the lower score of yachting (0.95),
where the vagaries of the weather make at least some dierence to the
outcome, and may partly explains why yachting is rated noticeably lower
than formula one (1.48): in both, the quality of the vehicle makes a dierence,
but the element of chance is greater in yachting. It is also likely to be a factor
in the lower score of shing (0.26).
The randomness factor also possibly explains the lower score of Fortnite
(−0.62) compared to the other two eSports, Counter-Strike (0.02) and League
of Legends (−0.51). Unlike in the other two eSports, in Fortnite the weapons
18 M. HEMMINGSEN
and supplies – integral to success in the match – are distributed randomly
throughout the map. Depending on chance, players may or may not have
easy access to essential resources that can have a signicant impact on the
outcome.
Third, events that have an aesthetic dimension, and that are therefore
judged rather than refereed – such as piano (−1.80), ballroom dancing (1.07)
and synchronised swimming (2.33) – tend to have lower scores. Perhaps
included in this group is equestrian dressage, though this activity was not
distinguished from equestrian activities more generally in the survey. The
idea here seems to be that the more that the aesthetic, judged elements
dominate the physical skill-based, refereed elements, the less that activity is
essentially a test of skill.
7
This is a criterion that comes in degrees: gymnastics
(2.68), gure skating (2.58) and diving (2.51) scored relatively high, synchro-
nised swimming (2.33) in the middle, and ballroom dancing (1.07) and piano
(−1.80) trailing far behind, tracking the increasing centrality of the aesthetic,
judged dimension of the activities.
Leisure
Being considered more of leisure activity than a serious, professional compe-
tition can potentially explain some of the survey results. This seems most
obvious in the cases of the ball sports with lower scores: all are associated
with leisure, with scores diering based on the extent to which this is true.
Given its association with relaxation and retirement, golf (2.49) is the obvious
example, but we can also note table tennis (2.58) – which is a common casual,
go-to activity for enjoyment with friends – and both pick up basketball (2.31)
and street football (2.41), which are otherwise more or less indistinguishable
from basketball (2.90) and soccer (2.93), but which were created explicitly for
casual enjoyment. We might also add bar sports, such as darts (1.08) and
billiards (1.26) to this list. Finally, it may be a contributing factor to the low
score of shing (0.26) and is perhaps a contributing factor to the scores of the
eSports as well.
Access
It also seems likely that there is an egalitarian dimension to sport. Those
activities that have socio-economic barriers to access received scores lower
than comparable activities. Obvious cases here are yachting (0.95), formula
one (1.48), horse riding (2.24) and especially golf (2.49) (given that it is
otherwise nearly indistinguishable from core sports). It is also possible that
the access factor explains golf’s lower score relative to the other leisure
activity of table tennis (2.58); and yachting’s lower score as compared to
formula one (1.48) (even if formula one vehicles cost more than yachts, at the
JOURNAL OF THE PHILOSOPHY OF SPORT 19
very least hopeful drivers can usually learn to drive on more accessible
vehicles, whereas a certain socio-economic status is required to even begin
learning how to sail a yacht).
8
The Inclusion of Leisure and Access
While features (1)-(3) are ‘built in’ to the activities themselves – they are
internal to their constitution – leisure and access are external; they are
about how people happen to relate to those activities. For instance, soccer
without essential physicality, or without being a test of skills, would no longer
be the same game. By contrast, while it just so happens that pick up basket-
ball is more of a leisure activity than a professional one, we can easily imagine
a world in which the very same game is taken seriously as a professional sport
without it changing it in any essential sense. Similarly, it just so happens that
golf has certain barriers to entry – often substantial green fees, club member-
ship costs, etc. – but it is quite possible to imagine a world in which that isn’t
true, but where golf is otherwise essentially identical.
One philosophical (i.e. non-social-scientic) question a core-periphery deni-
tion of sport must face, then, is whether a good denition of sport only includes
internal characteristics (M. Hemmingsen 2023b), or whether it is acceptable to
include external ones as well. If the former, then even if it turns out that leisure
and access are factors in explaining the extent to which various activities are
considered sports, this is merely a mistake in the popular understanding of
sport, borne of unclear thinking. If the latter, then leisure and access are truly
part of how we ought to dene sport, rather than simply how people just so
happen to do so. However, settling this question is outside the scope of this
paper, and arguably outside the scope of empirical research in general.
Conclusion
I have argued for an approach to dening sport that I have called a ‘core-
periphery’ model. This approach holds that there are certain ‘core’ sports that
possess all of the required features of the denition of sport, and do so to
a suciently robust extent. Other activities – those that lack one or more of
these features, or have one or more of the features to a reduced degree – may
still be considered sports, however they are less ‘sport-like’ than core sports,
and are located on the periphery.
In order to determine which features are required for an activity to be
considered a ‘core’ sport, an essential rst step is to engage in empirical
research into how the concept is understood in popular usage. The
results of such research are not the end of the discussion by any
means, as normative conceptual analysis is a matter for philosophical
discussion and debate. However, empirical research is required to ensure
20 M. HEMMINGSEN
that this analysis picks out the relevant concept, and also to help settle
issues that are not so amenable to philosophical analysis, such as the
degree to which the absence or attenuation of certain features makes an
activity less sport-like.
Notes
1. While the term ‘core-periphery’ may be familiar from, e.g. John Friedmann’s
‘core-periphery four-stage model of regional development’ (Friedmann 1966)
or Immanuel Wallerstein’s (1974) World Systems Theory of international rela-
tions, it has a distinct meaning in this context. In particular, the spatial element
of traditional core-periphery models in human geography and international
relations is metaphorical, and I am absolutely not suggesting – such as in World
Systems Theory – that there is any sense in which the ‘core’ of sport is in some
kind of exploitative relationship with those activities in the periphery.
2. Of course, for Part similarity comes down to psychological continuity and
connectedness, which is where the analogy to dening sport breaks down.
3. I am not suggesting that there is no place for philosophers in dening sport:
philosophers are still needed to tidy the concept of sport, to ensure that it is
coherent and consistent.
4. Again, I want to emphasise that I am engaged here in a demonstration of an
approach, rather than denitively arguing for a concrete denition of sport.
5. Cricket’s score is an odd case. Despite being very much the same kind of activity
as baseball, its score is much lower. However, compared to baseball or
American football, which are well known – even if not actively followed – in
countries without national teams or popular leagues, citizens of non-cricket
playing countries tend to know about cricket only in the vaguest terms, if at all.
I think it’s this ignorance that explains cricket’s comparatively low score. To test
this, I restricted the responses to only individuals from cricket-playing nations
(including Canada, where the sport is known, but could hardly be called out-
rageously popular), who would at least have some basic knowledge of the
game. When I did this, cricket’s score increased to 2.91, putting it decisively in
the top tier alongside baseball.
6. It is noteworthy that this criterion undercuts the status of activities that are
traditionally central to the Olympic and Commonwealth Games. Given that
being included in the Olympic Games is often taken to mean that an activity
is quite denitively a sport – to the point where inclusion in the Games has been
treated by some philosophers of sport to be a starting point for discussions
about dening sport – much more research is required to conclusively demon-
strate the relevance of the facilitative/constitutive distinction.
7. It is important to note that this is not saying in any sense that these activities are
unskilled; they may be as or more skilled than core sports. The point is that in
core sports, skill is the dominant thing that is being evaluated by the activity,
whereas in gure skating or diving, skill is at least somewhat decentred by the
evaluation of the aesthetic product of that skill.
8. While I didn’t have the data to undertake this analysis here, if access does
matter there is likely to be a negative correlation between respondents’ scores
and the median cost of golf memberships by country. Hopefully future research
will bear this out.
JOURNAL OF THE PHILOSOPHY OF SPORT 21
Disclosure statement
No potential conict of interest was reported by the author(s).
ORCID
Michael Hemmingsen http://orcid.org/0000-0001-9436-139X
References
Allen, B. 2013. “Games of Sport, Works of Art, and the Striking Beauty of Asian Martial
Arts.” Journal of the Philosophy of Sport 40 (2): 241–254. https://doi.org/10.1080/
00948705.2013.785418 .
Berman, M. N. 2013. “Sprints, Sports, and Suits.” Journal of the Philosophy of Sport
40 (1): 163–176. https://doi.org/10.1080/00948705.2013.785428 .
Friedmann, J. 1966. Regional Development Policy: A Case Study of Venezuela.
Cambridge: MIT Press.
Hallmann, K., and T. Giel. 2017. “eSports – Competitive Sports or Recreational Activity?”
Sport Management Review 21 (1): 14–20. https://doi.org/10.1016/j.smr.2017.07.011 .
Hemmingsen. 2021. “Code is Law: Subversion and Collective Knowledge in the Ethos
of Video Game Speedrunning.” Sport, Ethics & Philosophy 15 (3): 435–460. https://
doi.org/10.1080/17511321.2020.1796773 .
Hemmingsen, M. 2023a. “Movement Compression, Sports and eSports.” European
Journal for Sport and Society 1–18. https://doi.org/10.1080/16138171.2023.
2259176 .
Hemmingsen, M. 2023b. “Towards a Value-Neutral Denition of Sport.” Sport, Ethics
and Philosophy 1–16. https://doi.org/10.1080/17511321.2023.2260116 .
Hemphill, D. 2005. “Cybersport.” Journal of the Philosophy of Sport 32 (2): 195–207.
https://doi.org/10.1080/00948705.2005.9714682 .
Holt, J. 2016. “Virtual Domains for Sports and Games.” Sport, Ethics and Philosophy
10 (1): 5–13. https://doi.org/10.1080/17511321.2016.1163729 .
Holt, J. 2023. “Horses as Players in Equine Sports.” Sport, Ethics and Philosophy 17 (4):
456–464. https://doi.org/10.1080/17511321.2023.2189291 .
Howe, L. A. 2008. “On Competing Against Oneself, or ‘I Need to Get a Dierent Voice in
My head’.” Sport, Ethics and Philosophy 2 (3): 353–366. https://doi.org/10.1080/
17511320802475762 .
Howe, L. A. 2018. “Intensity and the Sublime: Paying Attention to Self and
Environment in Nature Sports.” Sport, Ethics and Philosophy 13 (1): 94–106.
https://doi.org/10.1080/17511321.2017.1388272 .
Howe, L. A. 2019. “Not Everything is a Contest: Sport, Nature Sport, and Friluftsliv.”
Journal of the Philosophy of Sport 46 (3): 437–453. https://doi.org/10.1080/
00948705.2019.1622126 .
Hurka, T. 2015. “On Judged Sports.” Journal of the Philosophy of Sport 42 (3): 317–325.
https://doi.org/10.1080/00948705.2015.1079137 .
Jenny, S. E., R. D. Manning, M. C. Keiper, and T. W. Olrich. 2016. “Virtual(ly) Athletes:
Where eSports Fit Within the Denition of ‘Sport’.” Quest 69 (1): 1–18. https://doi.
org/10.1080/00336297.2016.1144517 .
Kane, D., and B. D. Spradley. 2017. “Recognizing ESports as a Sport.” The Sport Journal
24.
22 M. HEMMINGSEN
Kobiela, F. 2018. “Should Chess and Other Mind Sports Be Regarded as Sports?”
Journal of the Philosophy of Sport 45 (3): 279–295. https://doi.org/10.1080/
00948705.2018.1520125 .
Krein, K. J. 2014. “Nature Sports.” Journal of the Philosophy of Sport 41 (2): 193–208.
https://doi.org/10.1080/00948705.2013.785417 .
Krein, K. J. 2015. “Reections on Competition and Nature Sports.” Sport, Ethics and
Philosophy 9 (3): 271–286. https://doi.org/10.1080/17511321.2015.1067249 .
Llorens, M. R. 2017. “eSport Gaming: The Rise of a New Sports Practice.” Sport, Ethics
and Philosophy 11 (4): 464–476. https://doi.org/10.1080/17511321.2017.1318947 .
Loy, J. W. 1968. “The Nature of Sport: A Denitional Eort.” Quest 10 (1): 1–15. https://
doi.org/10.1080/00336297.1968.10519640 .
Meier, K. V. 1981. “On the Inadequacies of Sociological Denitions of Sport.”
International Review of Sport Sociology 16 (2): 79–102. https://doi.org/10.1177/
101269028101600206 .
Meier, K. V. 1988. “Triad Trickery: Playing with Sport and Games.” Journal of the
Philosophy of Sport 15 (1): 11–30. https://doi.org/10.1080/00948705.1988.9714458 .
Morgan, W. J. 1977. “Some Aristotelian Notes on the Attempt to Dene Sport.” Journal
of the Philosophy of Sport 4 (1): 15–35. https://doi.org/10.1080/00948705.1977.
10654125 .
Paddick, R. J. 1975. “What Makes Physical Activity Physical?” Journal of the Philosophy
of Sport 2 (1): 12–22. https://doi.org/10.1080/00948705.1975.10654093 .
Part, D. 1986. Reasons and Persons. Oxford University Press. https://doi.org/10.1093/
019824908X.001.0001 .
Parry, J. 2019. “E-Sports are Not Sports.” Sport, Ethics and Philosophy 13 (1): 3–18.
https://doi.org/10.1080/17511321.2018.1489419 .
Schieman, K. 2016. “Hopscotch Dreams: Coming to Terms with the Cultural
Signicance of Sport.” In Dening Sport: Conceptions and Borderlines, edited by S.
E. Klein, 45–64. Lanham: Lexington Books.
Steenbergen, J. 2001. “The Double Character of Sport.” In Values and Norms in Sport,
edited by J. Steenbergen, P. Knop, and A. Elling, 33–56. Oxford: Meyer & Meyer
Sport.
Suits, B. 2005. The Grasshopper: Games, Life and Utopia. Peterborough, Ontario:
Broadview Press.
Wallerstein, I. 1974. “The Rise and Future Demise of the World Capitalist System:
Concepts for Comparative Analysis.” Comparative Studies in Society and History
16 (4): 387–415. https://doi.org/10.1017/S0010417500007520 .
Witkowski, W. 2012. “On the Digital Playing Field: How We ‘Do Sport’ with Networked
Computer Games.” Games and Culture 7 (5): 349–374. https://doi.org/10.1177/
1555412012454222.
JOURNAL OF THE PHILOSOPHY OF SPORT 23
