NATHALIE SINCLAIR

THE KISSING TRIANGLES: THE AESTHETICS OF

MATHEMATICAL DISCOVERY

ABSTRACT. Papert’s (1978) appeal to reconsider the power and possibilities of the
aesthetic in mathematics learning is often ignored in mathematics education research.
This paper begins with the premise, put forth by Dewey (1934), that the aesthetic struc-
tures many dimensions of inquiry and experience. In the same way that using particular
paintings, musical compositions, or even everyday experiences has been instrumental to
attempts by philosophers to understand the aesthetic dimensions of meaning and experi-
ence in artistic domains, I propose that analysing a particular encounter with mathematics
may help reveal the nature and role of the often nebulous responses of elegance, beauty,
and ‘fit’ to which mathematicians lay claim in their mathematical activity. To achieve this, I
draw on and adapt the defining features of the aesthetic character of experience set forth by
the aesthetician Beardsley (1982). This, in turn, sheds light on the role that aesthetics can
play in mathematical inquiry and experience, and provides initial categories and conjec-
tures that can be used to investigate the potential roles of aesthetics in mathematics learning
contexts.

KEY WORDS: aesthetic experience, dynamic geometry, investigation, motivation

Great mathematicians such as G.H. Hardy, Henri Poincaré, and Bertrand

Russell have attested to the aesthetic character of mathematical activity.
Many others, including contemporary ones such as Davis and Hersh
(1981), claim that aesthetic considerations are primary in mathematicians’
work. Based on its perceived function in the work of mathematicians,
a small number of educators have argued that aesthetic considerations
should be of greater importance in students’ learning of mathematics
(Dreyfus and Eisenberg, 1986; Higginson, 2000; Papert, 1978; Sinclair,
2001). However, it has been difficult to describe the roles aesthetics
can play in the mathematical learning process, largely because school
mathematics often inhibits rather than encourages aesthetic modes and
pleasures for students. Consequently, we know very little either about
the nature of children’s aesthetic responses in mathematics or how they
might contribute to both their cognitive and affective processes in the
mathematics classroom.
The existing literature on aesthetics and mathematics describes a range
of roles played by aesthetics, from motivating the choice of problems
– its motivational role (Penrose, 1974), to guiding the mathematician

International Journal of Computers for Mathematical Learning 7: 45–63, 2002.

© 2002 Kluwer Academic Publishers. Printed in the Netherlands.
46 NATHALIE SINCLAIR

to discovery – its generative role (Poincaré, 1908), to helping a mathe-

matician decide on the significance of a certain result – its evaluative
role (Tymoczko, 1993). However, both mathematicians and mathematics
educators have tended to isolate and give priority to its evaluative role
in making value judgements of mathematical products such as theorems
and proofs. In fact, as soon as aesthetics is mentioned in connection to
mathematics, most mathematicians are quick to assume that a reference is
being made to beautiful theorems and elegant proofs. Yet, a closer look
at the activity of mathematicians – their experiences doing mathematics
– suggests that the aesthetic also insinuates itself significantly during
the process of inquiry (prior to final evaluations) as the mathematician
selects, explores, struggles with, and shares a problem (viz. Davis, 1997;
Hofstadter, 1992). This aesthetic, being more personal and subjective,
stands in contrast to the formal aesthetic – that of mathematical elegance
and beauty – bandied about the professional community.
Apart from the three roles identifiable in the literature, there seems to
be another aesthetic dimension to mathematicians’ activities. When they
describe the pleasure they have at moments of discovery, the satisfaction
they get from pursuing a line of inquiry, and the transforming effect they
feel in coming to new mathematical understandings, they are attesting to
the aesthetic quality of their mathematical experiences. Following Dewey
(1934), there is aesthetic quality to an experience whenever there is a
coalescence into an immediately enjoyed qualitative unity of meanings
and values drawn from previous experience and present circumstances.
The mathematical inquiry can thus provide the means through which the
inquirer’s experience can achieve an aesthetic quality – a union of meaning
and value in which conclusions sum up and perfect the conditions that
lead up to them. Although the three roles of aesthetics discussed above
may contribute to such expeeriences, they are certainly not necessary
to them. For example, a mathematician’s experience may have aesthetic
quality without having to culminate in an elegant proof. Consider the case
of Andrew Wiles’ proof of Fermat’s Last Theorem. Few mathematicians
would call his proof elegant, perspicuous, or economical, yet no one who
has had the opportunity to hear Wiles speak of his journey would doubt its
consummatory nature.
I focus on Dewey’s (1934) notion of an aesthetic experience partly
for pedagogical reasons. The notion of an aesthetic experience is closely
related to the notion of an educative experience, one that not only inte-
grates the various aspects of experience – e.g., the cognitive, the affective,
the aesthetic – but transforms the self, broadening one’s perspective
through the acquisition of new meanings (Dewey, 1938). Dewey argued
 THE KISSING TRIANGLES 47

that such experiences are conducive to further growth, that is, they animate
students’ impetus to learn and they lead to independence, appreciation, and
development.
Dewey (1934) argues that in order to understand the aesthetic in its
“ultimate and approved forms,” one must begin with it in the raw, “in
the events and scenes that hold the attentive eye, arouse interest and
afford enjoyment” (p. 4). I suggest that in order to better understand what
initiates and constitutes the aesthetic character of experience in mathema-
tical activity, we need to examine particular, and “raw” examples of such
activity. Since it is difficult to monitor the thoughts and feelings of others
as they engage in mathematical activity, I examine my own experience
in order to develop a set of criteria for aesthetic experience. Since my
mathematical activity likely lies somewhere between that of a research
mathematician and that of a student in terms of prior knowledge and
past experience,1 I believe that my own aesthetic experiences in mathe-
matics might elucidate the nature of students’ aesthetic experiences in
mathematics.
In framing these criteria, I will draw on the writings of the Amer-
ican philosopher Monroe Beardsley (1982), who proposes the following
set of defining features of the aesthetic character of experience: object
directedness, felt freedom, detached affect, active discovery, and whole-
ness. This relatively late, more open-ended formulation provides for a
flexible consideration of the aesthetic dimensions of the experience in
mathematics. He uses particular examples from the artworld in order to
illustrate the features in question. Similarly, in what follows, I introduce
and illustrate Beardsley’s criteria as they manifest themselves throughout
the three stages of my mathematical discovery. The goal is to determine
the extent to which Beardsley’s criteria are applicable and explanatory in
this context of mathematical inquiry. In addition, I will be able to locate the
motivational, generative, and evaluative roles of aesthetics and investigate
how they operate within inquiry. This will provide some insight into the
different ways in which students might be either encouraged to or discour-
aged from aesthetically engaging in their own mathematical inquiries. I
discuss these pedagogical implications throughout the paper and in the last
section.

STAGE I: EXPLORING AND PATTERN-SEEKING

A colleague brought to my attention an interesting configuration that he

had constructed using the dynamic geometry software The Geometer’s
Sketchpad (Jackiw, 1991): start with any triangle ABC (shown in Figure 1)
48 NATHALIE SINCLAIR

Figure 1. Centres triangle DEF from triangle ABC.

and construct a square on each side of the triangle; find the centre of each
of the three squares and join them with three line segments (shown in bold)
to create the ‘centres triangle’ DEF. As he was describing it, I immediately
thought of the famous Napoleon’s theorem2 which begins with a similar
configuration. My colleague’s sketch showed a variation of it. The possi-
bility of a connection to a well-known theorem intrigued me, as did the
general appeal of the construction: would the random triangle ABC give
rise to a special derivative triangle, as in the case of Napoleon’s theorem –
embodying “order emerging from chaos” as Davis (1997) wrote?
Well, there wasn’t anything special about triangle DEF. But I noticed
that I could drag the vertices of the initial triangle ABC so that they
travelled inside and outside of the centres triangle DEF. Naturally, the in-
between case – the boundary condition – seemed of most interest and I
began to wonder under which conditions the vertex A would fall right on
segment DE: when would the vertex of the triangle ABC ‘kiss’ the centres
triangle? I dragged vertex A onto DE and measured the angle of vertex
A. With its accuracy of three significant digits, Sketchpad told me that
angle A measured 91.031◦ . Though an ‘ugly’ number, I was certain that
the angle, if I could get it more precisely on DE, would measure exactly
90◦ – betraying my own hope for a pretty result. But I needed to check out
this hypothesis which was, after all, based on no ‘logical’ evidence. First,
I had to figure out how to force that vertex to fall exactly on DE. But every
time I tried to drag A to that line, everything would move . . . The vertex
and the line were chasing each other around! I decided that if I could fix
the angle A at 90◦ – in other words, construct it such – I could see whether
A would then kiss segment DE. Fortunately, I could use Thales’ theorem:
if I construct a circle (shown in Figure 2) on the diameter BC, then I know
that each point on that circle will form a 90◦ angle with B and C. So, if I
define A on the circle, its angle measure will also become 90◦ .
 THE KISSING TRIANGLES 49

Figure 2. A kissing triangle. Vertex A lies on segment DE and on the circle.

Figure 3. Napoleon’s theorem configuration – equilateral triangles on a triangle.

Then came a moment of truth: once I moved the vertex A onto the
circle, it was kissing the segment DE (see Figure 2). And even when I
moved A around the circle, my triangles kept kissing: that was compelling
enough visually that I didn’t even bother to verify whether vertex A was in
fact right on segment DE or whether it just looked like it on the sketch. This
perfect fit made me wonder . . . Had I just found a particular relationship or
was there something bigger going on? I needed some more ‘data’ before
proceeding to a proof – to make sure that there was something worthwhile.
So I decided to try a slight variation. I constructed the Napoleon’s theorem-
like configuration with equilateral triangles on each side of the triangle
ABC instead of squares (as shown in Figure 3). Perhaps by looking at a
similar situation, I might start to see a pattern.

Motivation and Object Directedness: A Sense of Things Fitting Together

Beardsley describes the first feature of the aesthetic character of exper-
ience, object directedness, as “a feeling that things are working or have
worked themselves out fittingly” (p. 288) as one’s attention is fixed on
the qualities and relations of a perceptual or intentional field. In my case,
50 NATHALIE SINCLAIR

object directedness characterised my initial acquaintance with the forms,

patterns and structures of this configuration. Both the visceral feeling
generated by the joining of the vertex to the edge and the exactness of
the integer after my first construction gave me a strong visual sense of
‘fit.’ Making the connection to Napoleon’s theorem invited me to connect
to a recognised area of geometry; it was also gratifying to realise that a
configuration I had looked at and played with many times before held
even more treasures. In addition, the possibility of a framing structure was
emerging, from the specific case of the square to the general case of regular
polygons; this contributed to a deeper prescience of fit – I was anticipating
that relationships would emerge for a whole family of shapes.
This sense of fit motivated me to keep looking for relationships. Penrose
(1972) has identified choice of problem as the primary locus of the
motivational character of aesthetics. In this case, the initial appeal for the
configuration and the resulting observation was compelling enough to keep
me engaged enough to persevere with more technical and logical formu-
lations. After my first observation, I felt sure that I was about to discover
something.
Beardsley refers to both perceptual and intentional fields in object
directedness. Whereas the perceptual might apply to, in this case, the
visual or kinaesthetic aspects of the problem’s configuration, the inten-
tional applies to the more narrative, temporal aspects. There is a sense in
which I apprehended the relationships in the configuration in a particular
order, as if I was unravelling a story. As with reading a novel, I wanted
to find out what would happen next, what theme would emerge from the
sequence of ideas. Rather than beholding a fixed mathematical truth, which
can be apprehended at once, like a painting, this felt more like gradual
revelation through exploration.
During this first stage, I could sense things fitting together. Indeed, I
was guided by that very sense in my conjectures. I felt confident that the
awkward initial angle measurement would turn out to be an integer and
that the existence of this integer would be evidence of some emerging
relationship. This might have been the generative character of aesthetics at
work: the aesthetic choice, one of ‘neatness,’ was to see this number as 90◦
and to pursue the implications. I turned the decimal number into a number
that is not only an integer, but one that appears frequently in Euclidean
geometry. This action at once suggested and confirmed my belief that there
existed a discernible relationship and provided shape to the conjectures I
would make as to the nature of this relationship. Absent this turn to 90◦ , I
might have dismissed the kissing state as uninteresting or concluded that
the angle measurement would not be involved in defining a relationship –
 THE KISSING TRIANGLES 51

Figure 4. One case produces kissing triangles.

I might even have abandoned my pursuit had the angle measured 51.031◦
instead. Similar aesthetic choices are involved when mathematicians make
conjectures based on the discernment of a special mathematical object or
relation such as square numbers, parallel lines, or striking patterns (viz.
Silver and Metzger’s (1989) prime factorisation example).
This stage of the activity was an important one both aesthetically
and motivationally. I became increasingly attracted and motivated to
understand the fit – to look for a heuristic or proof that would explain
it.

STAGE II: MAKING INTELLIGIBLE

Having constructed the equilateral-triangles configuration, I couldn’t wait

to see what the kissing angle would be. Once again, Sketchpad gave me a
‘messy’ angle when I placed vertex A on segment DE, but it was pretty
close to 120◦ and I felt a little rush of hope. As before, I wanted to
verify my conjecture by constructing A to measure 120◦ . I realised that
this time Thales wouldn’t help. Anxious to forge forward, I constructed
one specimen triangle that would satisfy my need: if one angle had to be
120◦ then I could just construct an angle at B to measure 20◦ and an angle
at C to measure 40◦ . This 120-20-40 triangle would be clearly scalene
and therefore wouldn’t introduce any artificial relationships (see Figure 4).
When I moved vertex A to the 120◦ vertex X of my 120-20-40 triangle, my
triangles kissed! I felt sure that I was on to something.
But I still had two lingering problems: the first was conceptual – to
understand the relationship I was closing in on. The second was more
technical – to figure out how to construct an arc of a circle that allow me
to look at all triangles with  CAB = 120◦ , not just the specific 120-40-20
case. I had been able to construct the general case triangle for the squares
case and I became intrigued by whether I could do it for this case, even
52 NATHALIE SINCLAIR

Figure 5. The regular hexagon case.

though it might not bring me any closer to solving my first problem. And
if I succeeded, I would have the pleasure and verification of seeing my
kissing vertex in action along the arc. As I pondered these two problems,
I constructed another variation which I hoped would verify my emerging
relationship between the outer polygon and the kissing angle (so far I knew
that equilateral triangles have a kissing angle of 120◦ and that squares have
a kissing angle of 90◦ ). This time I constructed a regular pentagon on my
triangle ABC (see Figure 5). I soon discovered that the kissing angle for the
pentagon was 72◦ . So, now my kissing angle theorem seemed obvious –
the kissing angle is just 360◦ divided by the number of sides of the regular
polygon: 360◦ /n.
It was while I was attending to the 60◦ angle in the regular hexagon
case that I realised I could solve my second, technical problem . . . I knew
that I needed an arc through BC such that any point on that arc would
subtend the chord BC with a constant angle (specifically, a 120◦ angle) – I
needed the family of triangles BX C on BC for which X was 120◦ . I had
racked my brain to remember whether there was some kind of procedure
that, given the chord BC, would let you find the requisite circle, some
variation on Thales perhaps? But to no avail. So I went for a more ‘mech-
anical’ solution: I could construct the locus of points X – which would
form an arc – such that angle BX C measured 120◦ (see Figure 6). That
required that the angles at B and C add up to 60◦ (the hexagon angle!).
This construction was one which I knew how to execute in Sketchpad! I
could use a parameter. I constructed an angle FDE of measure 60◦ , and
then a parameter point X along the arc FE, so that  XDE +  FDX = 60◦ .
Next, I constructed a ray from point B such that the ray would form an
angle to BC equivalent to  XDE and a second ray from C such that the
second ray would form an angle to BC equivalent to  FDX. I found the
intersection of the two rays, X . By construction,  BX C measured 120◦ .
Therefore, moving parameter X along the arc FE moved X along an arc
 THE KISSING TRIANGLES 53

Figure 6. Using a parameter construction: each point X on the arc creates a 120◦ angle
BX C.

describing all the possible triangles such that  BX C measured 120◦ . I
used Sketchpad’s locus command to construct this arc BX C. The beauty
of this method was that it would work for any angle. It didn’t seem like
a pretty geometric method, but it was powerful. The culmination of this
construction was to animate point A along the arc BX C and to watch how
point A maintained its kiss.
I felt a certain amount of satisfaction at having resolved my technical
problem, so I returned with renewed confidence to my first problem. It
was time to pull out a notepad and return to the square case, which was
so familiar. Once I drew the configuration (see Figure 7), I noticed several
new properties. First, when vertex A was on segment DE, the triangle ABC
seemed to be congruent to triangle NAM. In fact, if I could prove that
congruence then it would follow that the kissing angle A had to be 90◦ .
After several false starts – mainly looking for similar triangles – I noticed
the quadrilateral CMNB and the fact that segment DE seemed to bisect
both MC and NB. In fact, this was obvious (E is the midpoint of diagonal
MC since it is the intersection of the diagonals of the square on AC and
similarly, D is the midpoint of diagonal NB). Therefore CEDB is congruent
to MEDN (since EM = EC; ND = DB; DE = DE, and  NDA =  BDA =
 MEA =  CED = 90◦ ). So MN is equal in length to CB. And thus, by the
SSS rule, · NAM is congruent to · ACB, and so  NAM =  CAB. Since
 MAN +  CAB = 180◦ (because  MAC =  NAB = 90◦ ), we have that 2
CAB = 180◦ and  CAB = 90◦ . My proof for the squares case of the kissing
angle theorem was complete!
It was complete, but particular to the squares case.3 I liked the sequence
of relationships that led to the final step of the proof. It also highlighted
properties of the configuration that I hadn’t attended to in my explorations
but that started to explain how the outer polygon and its centre point inter-
acted with the triangle ABC and its kissing vertex. And I had a vague idea
of how I could explain the other configurations. Yet it still felt like I was
54 NATHALIE SINCLAIR

Figure 7. Towards a proof of the kissing angle theorem.

missing something simpler. And I had certainly only proved one case of
the kissing angle theorem.

Proof, Active Discovery and Felt Freedom

Beardsley describes another feature of the aesthetic character of experi-
ence, active discovery, as a highly cognitive act of:
exercising constructive powers of the mind, of being challenged by a variety of potentially
conflicting stimuli to try to make them cohere: a keyed-up state amounting to exhilaration
in seeing connections between percepts and between meanings, a sense (which may be
illusory) of intelligibility. (p. 289)

He remarks that this criterion is the closest to mathematicians’ experi-

ences during problem-solving. Indeed, I had a sense of having insight into
connections and organisations; I enjoyed the emerging intelligibility of the
relationships I had conjectured and was sure were true. The proof put into
focal awareness new relationships and made me feel like I had understood,
made more intelligible, the essential properties of the kissing triangles. It
hardly mattered what regular polygon one started with because it was only
involved through its centre, which, for the squares case, held the key to the
similarity argument in the proof.
My sudden realisation about how to execute the technical construction
came after I had left the problem momentarily to do something less cogni-
tively taxing. Hadamard (1945) details reports of mathematicians having
this experience during a problem-solving process and attributes either to
a freshness of mind, a forgetting of previous false leads, or the result
of the mind’s subconscious work. Poincaré (1908) makes a more ambi-
tious claim; he refers to this stage of problem-solving as an incubation
period during which numerous combinations of elements are generated
in the subconscious and then passed through an aesthetic sieve back to
 THE KISSING TRIANGLES 55

the conscious mind. This claim has found tentative support in recent
neuroscientific studies (Damasio, 1994).
I would like to highlight a difference between the two stages of my
journey. The first one carried with it a sense of mystery, a discovery of a
new, unformalised property. During the second stage, I moved from the
immediate properties and mysteries and started working in the language
of mathematics. This too holds a certain appeal, to be able to move from
one mode of expression to another, to see how the formal mode crystallises
properties and leads you through an almost inevitable sequence of steps.
The two stages thus might even exemplify aesthetic preferences: while
some prefer certainty, others, such as André Weil prefer mystery or ‘just’
intelligible phenomena:
Nothing is more fruitful – all mathematicians know it – than those obscure analogies, those
disturbing reflections of one theory on another; those furtive caresses, those inexplicable
discords; nothing also gives more pleasure to the researcher. The day comes when this
illusion dissolves: the presentiment turns into certainty; the yoked theories reveal their
common source before disappearing.

Contrary to the emphasis in the mathematics literature on the proof as

the ultimate aesthetic object, my own aesthetic experience was only partly
about proof. It was more importantly about seeing new relationships, fitting
them together, using intuitions and conjectures, creating patterns. In fact, I
could even say that the very combination of the intuitive, figurative work
of the first stage and the more logical, literal work of the second stage
provided a sense of fit of my own emerging intelligibility. I still had a
nagging feeling that I did not find an elegant proof, evidence of a powerful
evaluative aesthetic manifesting itself. Perhaps I didn’t like the fact that
I had to use the derivative quadrilateral CMNB in my proof. Perhaps I
wanted to see more clearly how the special case proof could extend to
the general case. Perhaps the proof was simply not obvious enough. This
aesthetic unease often motivates mathematicians to search for better proofs
– as it ultimately did in my case (see Appendix).
Beardsley describes the criterion of felt freedom as a “sense of release
from the dominance of some antecedent concerns about past and future,”
and a “sudden dropping away of thoughts and feelings that were problem-
atic, that were obstacles to be overcome.” I certainly felt a release from the
tension of having to prove this result. I experienced two distinct releases
in the second stage. The first was when I figured out how to construct
the arc using my mechanical method. The second was when I found my
proof for the kissing angle. The first release was more pleasurable, and in
retrospect I can pinpoint four reasons for this. First, figuring out how to
effect the construction I needed contributed more to convincing myself of
56 NATHALIE SINCLAIR

the relationship than finding a proof did. Second, I had spent much time
trying to recall some hidden property that would allow me to make the
construction using circle properties and felt a sense of relief both at the
fact that remembering a remote property wasn’t important and that using
a method I often use in other situations was applicable in this context.
Third, I am more confident in general with my problem-solving abilities
in Sketchpad than my proof finding abilities in geometry. Although a lack
of confidence might induce more tension and thus a higher relief, I believe
that part of the feeling of release is the build-up of positive tension so
that the sense or release is rather more integrative in the end. The fourth
reason is that although I was very pleased with my solution to the first
construction problem, I felt that the solution to my second proof problem
wasn’t as revelatory or simple as it could be.
I am intrigued by the obstacles that the felt freedom criterion might
present to students. First, there is the confidence element: do I believe that
I am capable of finding a solution? If not, it is unlikely that I will allow
myself to experience the build-up of tension. Second, is there an external
criterion of rigour or acceptability that I am not sure how to attain? If so,
then the antecedent concerns are external in nature and unlikely to provide
the aesthetic sense of release.
I also experienced a different kind of rhythmic release from the domi-
nance of my concern about finding a proof by going back to Sketchpad on
occasion to continue my constructions of each of the hexagonal, octagonal,
and decagonal configurations. It was peaceful and comforting to play with
these configurations, compared to the tension involved in looking for a
proof.

STAGE III. CONNECTING AND PROVOKING

I thought there must be more going on with this rich situation. Finding one
relationship provoked me to look for more. For instance, I wondered how
many vertices of the triangle ABC I could get on the centres triangle, or
even inside the centres triangle. Using my established result that the kissing
angle is 360◦ /n where n is the number of sides of the outer polygon, I could
easily establish that the number of vertices that can get inside the centres
triangle is m, where m is the largest number such that m* (360◦ /n) is less
than 180◦ . This result, simple as it was, crowned my exploration.
 THE KISSING TRIANGLES 57

Generativity and Wholeness

Beardsley describes his fifth criterion of wholeness as “a sense of inte-
gration as a person . . . and a corresponding contentment, even through
disturbed feelings, that involves self-acceptance and self-expansion”
(p. 289). The three stages I described are interesting to consider in terms
of their wholeness. The first was one of initial appeal providing motivation
and excitement but also remaining raw and open. The second stage was
the taming one, both of my problem and my thinking process – I attempted
to turn the problem into a formal entity, not only for the sake of formal-
ising but to crystallise and to express. The last stage gave me a sense of
the generativity of the problem, of the potential connections and further
relationships – for instance, what other shapes could I try besides regular
polygons? In this stage, I returned to a sense of freedom and playfulness.
Moreover, I had created a little universe, a theme within geometry that I
was free to manipulate and vary as I wished. The sense of integration thus
has two intertwined dimensions: on one level I experienced as a whole the
rhythm of tension and freedom, on both an intellectual and emotional level;
and, on another level, I claimed ownership of a little universe that reflected
my own position in relation to the mathematics. In fact, Beardsley stresses
not only the coherence of the elements of the experience but of the self.
There is a certain wholeness about connecting my visual play with my
analytic formalising, about turning what’s in my head to something that I
can express and share with others, but mostly perhaps, about the very act
of creating and discovering something.

The Criterion of Detached Affect

The other criterion that Beardsley posits is that of detached affect, which
he describes as “a sense that the objects on which interest is concentrated
are set a little at a distance emotionally” (p. 288). I find this criterion harder
to match to my own aesthetic experience in mathematics.4 The element of
detachment has often been debated and is even considered central to the
nature of aesthetic experience by some aestheticians. It is thought that the
aesthetic of detached affect emerges from the subtle interplay between the
real and the fictional, the authentic and the artificial. It is seen to account
for, at least in part, the aesthetic experiences we derive from the ugly, the
violent, and the tragic.
Dewey (1934) provides another perspective on the “detached affect”
phenomenon in his discussion of aesthetic response to the purportedly ugly
and tragic. He proposes that when a particular subject matter is removed
from its practical context, it enters “into a new whole as an integral part of
it” (p. 96) so that the former ugliness is contrasted with a new quality. It
58 NATHALIE SINCLAIR

acquires a new expression on account of its new relationships. Instead of

Beardsley’s “detached affect”, I therefore propose a criterion more akin to
Dewey’s notion which I will refer to as extraction: the removal of some-
thing from its usual setting so that it becomes part of an expressive whole.
This begins to capture aspects of my experience with the kissing triangles.
Those inert mathematical objects of triangles, angles and polygons, were
removed from their usual setting and configured into a new one which
bound them together as a unit, and in so doing, bound me to the unit. What
is the usual setting of triangles, angles and polygons? From my point of
view, triangles are set by typical categorisations (equilateral, isosceles,
and scalene), or properties (measurement formulae, special centres and
lines), or relationships (congruence, similarity). The new setting bounded
triangles with other polygons, with inside- and outside-ness, with a derived
triangle, and with kissing. The latter feature of the new setting suggests the
way in which I became part of this expressive whole.

Revisiting the Criteria

In the realm of mathematics, it would seem natural for the third criterion,
that of active discovery, to be a necessary condition. However, in terms
of my own experience, meeting the third criterion depended heavily on
activating the first one of object directedness. I described object directed-
ness largely in terms of its motivational role in both drawing me to the
problem and initiating pattern seeking. What if I had stopped after stage
one, thus meeting the criterion of object directedness but not that of active
discovery? I would argue that it is the sense of active discovery that makes
possible both felt freedom and wholeness. I have often aborted investiga-
tions after stage one and felt a lingering lack, an incompleteness. I propose
that the third criterion is necessary but unlikely to occur, perhaps at varying
degrees for different people, without the first. The criterion of wholeness
seems to act like a wrapper; it adds a crucial dimension of personalisation
to the experience, a type of awareness that makes one cognisant of their
own self in relation to the mathematics. Without this criterion, though it
infuses to a certain extent – and for some more than others – the first and
third, one can imagine an almost mechanical experience.
I now consider the two criteria of felt freedom and extraction again. In
terms of confidence and autonomy, it would seem that felt freedom acts as
a condition for and physical manifestation of active discovery. It also seems
to carry the majority of the affective weight of the five criteria: the sense
of release, relaxation, harmony that are all pleasurable. This pleasure both
instigates and reveals or expresses the cognisance of fit afforded by active
 THE KISSING TRIANGLES 59

discovery. It might in fact be necessary in terms of alerting the observer to

his or her own aesthetic response.
The criterion of extraction seems the most unpredictable in terms of
students’ aesthetic experiences in mathematics. My description of extrac-
tion with kissing triangles exemplifies this to a certain extent. I created a
new setting for the rather dry, and perhaps unappealing (to many), objects
of mathematics which allowed me to see their liveliness – allowed me even
to anthropomorphise – and activate their properties and relationships. I
am not convinced that this process is as important to a mathematician as
it might be to a student for two reasons. The first concerns the dulling
experience of mathematics that students often encounter in school, one
that is often void of any sense of beauty. Many of them are prone to
see mathematics as ugly and perhaps tortuous. The process of extrac-
tion for them would thus be crucial, paralleling the effect described by
Dewey above. The second is that the process of extraction might also
focus students’ attention on new properties and relationships that might
otherwise be overlooked.

Pedagogical Implications
What pedagogical insights can this introspective analysis provide? It is not
prudent to draw conclusions or make specific recommendations based on
a single case of mathematical inquiry. However, under the assumption that
a goal of mathematics education is to provide students with opportunities
to have similar experiences – i.e. learning experiences that are intrinsically
satisfying – then this analysis points to some of the enabling conditions that
we can strive to meet in designing mathematics learning environments.
Many of the enabling conditions that we can identify from this analysis
are those which mathematics educators have long been promoting: the
opportunity to ‘play around’ and to explore (Hawkins, 2000; Papert, 1980);
the possibility of posing personally and epistemologically interesting prob-
lems (Brown and Walter, 1983; Brown, 1996); the time to develop interest
in and intimacy with a mathematical situation (DeBellis and Goldin, 1999)
– and the possibility of leaving it when stuck; the opportunity to make
connections with previous knowledge and past experience; the chance
to be intrigued by, surprised by, or attracted to a mathematical situation
(Goldin, 2000); and, the chance to build confidence in one’s abilities.
However, this analysis also provides a description of how these
enabling conditions function in the process of mathematical inquiry, and
how they manifest themselves in particular, human contexts. For example,
making a connection with Napoleon’s theorem motivated my inquiry by
lending a sense of significance to it. Seeing Napoleon’s theorem configura-
60 NATHALIE SINCLAIR

tion under a new light both surprised and delighted me, much as one
is surprised and delighted when they realise how they have overlooked
something which in retrospect seems obvious. The attractiveness of the
configuration invited my full attention, drawing me in to notice certain
details and relationships. This analysis highlights the important role that
aesthetic motivation plays in inquiry, not only at the affective level, but
also, in shaping inquiry by influencing the discernment and selection of
features in a situation and in directing the thought patterns of the inquirer.
Another enabling condition was the time I had to play around, to
explore, and to experiment. I had a chance to connect and use my past
experience and knowledge to discover new ideas – without those partic-
ular discoveries being the predetermined goal of my activity; I was free
to notice properties that followed from my own curiosities and tendencies.
From a student’s perspective, I was seeing ‘already learned’ properties and
relationships in a new context, allowing me to bring together what I knew
with what I could subsequently learn. Might students have better access
to aesthetic experiences if they were given the time and freedom to ‘go
more deeply’ with what they have already learned rather than being pushed
higher and higher to new skills and concepts?
In terms of the enabling condition of developing interest and intimacy,
Sketchpad’s capacity to allow for approximations (dragging things ‘close
to’ without having to explicitly construct them so) offered accessibility,
with an initial entry point that my mathematical imagination alone might
not have found. Its dynamic nature often provided striking effects that my
mental patterns would not have been able to simulate. Finally, having the
chance to struggle through and solve one problem, unrelated as it was to
my original inquiry, gave me the confidence to keep struggling with the
subsequent problem; it even provided me with a taste of what it would feel
like to solve subsequent problems.
Finally, in terms of research, this analysis provides insight into the
criteria that we can look to identify when trying to recognise such exper-
iences in students – this is crucial if we are to try to determine how
such experiences might affect their cognitive and affective development.
In addition, and perhaps most importantly, this analysis underscores the
fact that aesthetic responses are not only useful for determining the signifi-
cance or value of a mathematical result; rather, they help shape and drive
mathematical inquiry – making it both possible and fruitful – and, they
contribute to the feelings of personal satisfaction and pleasure that stimu-
late continued growth and appreciation. This should provide added impetus
for continued research into the roles of aesthetics in mathematics learning.
 THE KISSING TRIANGLES 61

ACKNOWLEDGEMENTS

I wish to thank Bill Higginson for his helpful encouragement and sugges-
tions in writing this paper, SSHRC for supporting my research, and my
reviewers for their very helpful comments.

NOTES
1 I earned a Master’s degree in mathematics but I have never engaged in professional
research in mathematics.
2 Napoleon’s Theorem states that the triangle formed by joining the centres of three equi-
lateral triangles constructed on the sides of any triangle will also be equilateral.
3 In fact, I found a general proof of the kissing angle theorem which highlights an element
of symmetry that this special case proof obscures. It is included in the appendix.
4 Bill Higginson (personal communication) has suggested that mathematicians might
experience the sense of detached affect in their constructions of proofs which are so often
stripped of the situations and examples to which they apply or the personal commitments
and attractions that formed them.

APPENDIX

Proof for the kissing angle theorem:

Let ABC be any triangle.

Construct regular n-gons on each side of ABC and let DEF be the ‘centres’
triangle.
Assume that vertex B lies on segment DE.
Since BD is a line of symmetry for polygon on AB and BE is a line of symmetry
for polygon on BC, then DBE acts as a line of symmetry through hexagon
ADNMEC.
62 NATHALIE SINCLAIR

Therefore  ABC = 180◦ − ( ABD +  CBE)

= 180◦ − 2* ABD
= 180◦ − (180◦(n-2)/n)
= 360◦ /n

REFERENCES

Beardsley, M.C. (1982). The Aesthetic Point of View. Selected Essays. Ithaca, NY: Cornell
University Press.
Brown, S. (1996). Towards humanistic mathematics education. In A.J. Bishop et al. (Eds),
International Handbook of Mathematics Education (pp. 1289–1321). Dortrecht, The
Netherlands: Kluwer Academic Publishers.
Brown, S. and Walter, M. (1983). The Art of Problem-Solving. Philedelphia: The Franklin
Institute Press.
Damasio, A. (1994). Descartes’ Error: Emotion, Reason, and the Human Brain. New York:
Avon Books.
Davis, P. and Hersh, R. (1981). The Mathematical Experience. Boston: Birkhauser.
DeBellis, V. and Goldin, G. (1999). Aspects of affect: Mathematical intimacy, mathema-
tical integrity. In O. Zaslovsky (Ed.), Proceedings of the 23rd International Conference
of the International Group for the Psychology of Mathematics Education, vol. 2
(pp. 249–256). Haifa, Israel.
Dewey, J. (1938). Experience and Education. New York: Collier Books.
Dewey, J. (1934). Art as Experience. New York: Perigree.
Dreyfus, T. and Eisenberg, T. (1986). On the aesthetics of mathematical thought. For the
Learning of Mathematics 6(1): 2–10.
Goldin, G. (2000). Affective pathways and representation in mathematical problem
solving. Mathematical Thinking and Learning 2(3): 209–219.
Hadamard, J. (1945). The Mathematician’s Mind: The Psychology of Invention in the
Mathematical Field. Princeton, NJ: Princeton University Press.
Hawkins, D. (2000). The Roots of Literacy. Boulder, CO: University Press of Colorado.
Higginson, W. (2000). Amusing about aesthetics and mathematics. In J. McLoughlin
(Ed.), Proceedings of the 2000 Annual Meeting. Canadian Mathematics Education Study
Group, Topic Group A. St. John’s: Memorial University of Newfoundland.
Hofstatder, D. (1996). Discovery and dissection of a geometric gem. In J. King and D.
Schattschneider (Eds), Geometry Turned On: Dynamic Software in Learning, Teaching,
and Research. Washington, DC: MAA.
Jackiw, N. (1991). The Geometer’s Sketchpad. Berkeley, CA: Key Curriculum Press.
Papert, S. (1980). Mindstorms: Children, Computers, and Powerful Ideas. Brighton, UK:
Harvester Press.
Papert, S. (1978). The mathematical unconscious. In J. Wechsler (Ed.), On Aesthetics and
Science (pp. 105–120). Cambridge, MA: MIT Press.
Penrose, R. (1974). The role of aesthetic in pure and applied mathematical research. The
Institute of Mathematics and its Applications 7/8(10): 266–271.
Poincaré, H. (1908/1956). Mathematical creation. In J. Newman (Ed.), The World of
Mathematics, vol. 4. New York: Simon and Schuster.
 THE KISSING TRIANGLES 63

Silver, E. and Metzger, W. (1989). Aesthetic influences on expert mathematical problem

solving. In D.B. McLeod and V.M. Adams (Eds), Affect and Mathematical Problem
Solving (pp. 59–74). New York: Springer-Verlag.
Sinclair, N. (2001). The aesthetic is relevant. For the Learning of Mathematics 21(1): 25–
33.
Tymoczko, T. (1993). Value judgements in mathematics: Can we treat mathematics as an
art? In A.M. White (Ed.), Essays in Humanistic Mathematics. Washington, DC: MAA.

Faculty of Education
Duncan McArthur Hall
Queen’s University
Kingston, ON
K7L 3N6
E-mail: nathsinc@educ.queensu.ca