





































Building Plans as Natural 
Symbols

Rafael De Clercq

At present, there are no hotly debated topics 
in the philosophy of  architecture. One could 
take this as evidence of  the dreariness of  the 
discipline, but one could equally see it as a sign that 
the discipline has not been caught up in its own 
internal dialectic and thereby lost touch with issues 
that matter outside of  philosophy. After all, when a 
(philosophical) discipline lacks a compelling internal 
dialectic, there is every reason for researchers in 
the field to look beyond its boundaries, to other, 
neighboring fields. In the case of  the philosophy of  
architecture, these neighbors include architectural 
history, architectural theory, architectural criticism, 
and, last but not least, the practice of  architecture 
itself. It should come as no surprise, then, that the 
topic of  my paper is an entirely self-selected one 
from the field of  architectural history and theory. 
More specifically, I will focus on a claim about 
building plans that can be found in the work of  
architectural historian Carroll William Westfall. The 
claim has received little attention, perhaps because 
it is considered implausible, or because it is made 
by someone whose taste in philosophy (Aristotle) 
and architecture (Jefferson) seems all too classical. 
Whatever the case may be, in what follows I will 
try to argue that the claim is not implausible. I will, 
however, give my own interpretation to it, one that 
may not match entirely what Westfall had in mind.1 
Still, the claim is supposed to remain sufficiently 
strong under the proposed interpretation to make 
it far from trivially true.

there are 
buildings 

belonging to 
disparate 

functional kinds 
(e.g., ‘temple’, 

‘dwelling’, ‘shop’) 
whose plans are 
natural symbols 

of the activities 
accommodated 

by those 
buildings 

“

”



AP . vol 1 . No 1 . 2014

62

De
 c

le
r

cq

I.

Westfall advanced his thesis about building plans in Architectural 
Principles in the Age of  Historicism, a book co-authored with Robert Jan van 
Pelt in 1991.2 Here are some relevant quotations from Westfall’s chapter 
on building types:3 

Here it is argued that a building imitates a type which is timeless. The type 
provides a symbol of  the purpose which the building embodies… The particular 
building is a conventional sign within which is embedded the natural symbolism 
of  the type (van Pelt & Westfall 1991, p. 156). 

In this sense, [the type] is a natural symbol of  the political purpose it 
accommodates. In both the most simple and most complex way, then, the type’s 
character is embodied in its plan diagram, and that plan diagram is about a 
purpose… (Ibid., p. 160)

Each purpose has a corresponding building type with a characteristic plan 
diagram. The type embodies the continuity between the present and the past. It is 
that which endures because it is true. Because it is true, it exists outside the time 
marked by history, and it has symbolic value—it symbolizes some particular 
political purpose (Ibid., p. 166).

The claim I wish to extract from the above quotations can be stated 
as follows: there are buildings belonging to disparate functional kinds 
(e.g., ‘temple’, ‘dwelling’, ‘shop’) whose plans are natural symbols of  the 
activities accommodated by those buildings. It may not be immediately 
clear why this claim is extracted from the quotations. First of  all, in the 
quoted passages, Westfall is not saying that activities are symbolized; he 
says that purposes are symbolized. However, activities and purposes are 
not clearly distinguished in van Pelt & Westfall 1991 (e.g., on pp. 156, 161-
2). Moreover, as we will see in section III, the explanation of  the symbolic 
relation may be roughly the same regardless of  whether purposes or 
activities are symbolized. In other words, the plausibility of  the central 
claim does not depend on which of  the two interpretations is preferred.4 

Another interpretative issue arises from the fact that Westfall does not 
say that building plans are symbols; he tends to say that building types are 
symbols. However, since, according to the second quotation, “the type’s 
character is embodied in its plan diagram” it does not seem too extravagant 
to take him as claiming that building plans are symbols or, if  one prefers, 
that buildings symbolize by virtue of  their plan. Indeed, more recently, 
Westfall writes that (particular) buildings “based on” building types are 



63

isparchitecture.com

natural symbols (unpublished, p. 14).

In any case, my interpretation of  the claim 
will henceforth focus on two expressions: ‘plan’ 
and ‘natural symbol’. My discussion of  the 
claim’s significance—in particular, how it differs 
from similar claims—will have to wait until the 
conclusion, when a clear interpretation has been 
provided.

II.

The plans that are supposed to serve as natural 
symbols are not diagrammatic representations of  
horizontal sections (these are “conventional signs”), 
but the horizontal sections themselves. Moreover, 
they are the horizontal sections of  particular 
buildings, for example, a horizontal section of  
the ground floor of  the Houses of  Parliament in 
London. However, if  we follow Westfall, these plans 
or sections are able to serve as natural symbols only 
because they exemplify a certain abstract shape; in 
other words, because they are tokens of  a certain 
type. Westfall identifies six basic types: the tholos, 
the temple, the theatre, the regia, the dwelling, and 
the shop. (For our purposes, we can regard a basic 
type as one whose symbolic significance does not 
derive entirely from its being an instance of  another, 
more general type.) The activities symbolized by the 
tokens of  these types are, respectively, venerating, 
celebrating, imagining (or “aspiring”), governing 
(or “exercising authority”), dwelling, and sustaining 
(or “trading”). The diagrams in figure 1 represent 
the abstract shapes the tokens have in common in 
virtue of  being tokens of  the same type.

When introduced in this way, the list may strike 
one as somewhat arbitrary, and indeed Westfall does 
not explain in detail how he arrived at it. He says little 
more than the list is based on, “[e]xperience with 
historical building and reflection about the historic 
within that experience as well as the knowledge 
available to us through our life in the present” (p. 

What 
distinguishes 

natural from 
non-natural 

symbols is 
the way the 
connection is 
established 

between the 
symbol and what 

it stands for

“

”



AP . vol 1 . No 1 . 2014

64

De
 c

le
r

cq

156). Nonetheless, it seems to me that the list cannot be rejected out of  
hand as based on someone’s parochial (say, Western) experience. This may 
be evident in the case of  the dwelling and the shop, but even the peculiar-
looking tholos is exemplified by religious structures around the world such 
as mosques and Chinese altars and pagodas. The apparent universality 
of  the types is of  course due to the fact that they are very general and 
leave many features undetermined, for example, dimensions of  rooms, the 
placing of  windows and doors, etcetera.6 However, such indeterminacy 
does not empty Westfall’s list of  significance as long as the shapes, and the 
corresponding activities, are recognizable and recognizably different from 
one another.7 That is enough for Westfall’s list to be refutable. Moreover, 
Westfall leaves open the possibility that a different (and presumably, equally 
correct) list can be produced on the basis of  a “different understanding 
of  the way political life is given conventional form in constitutions and a 
different interpretation of  the activities that constitute political life” (p. 
157). However, in what follows, I do not want to defend even the thesis 
that Westfall’s list is a correct enumeration of  basic types. All I want to 
argue is that some such list (perhaps a much longer one) may be correct. 
That is enough for the claim under investigation to be plausible or at least 
worthy of  further investigation.

The claim extracted from Westfall’s chapter makes reference only to 
horizontal sections. In principle, one could also take vertical sections into 
account. The word ‘plans’ in my formulation of  the main thesis would 
then have to be replaced by something like ‘horizontal and/or vertical 
sections’. This should not affect (what I take to be) the substance of  the 
claim, although it may make it more plausible. In what follows, however, 
I will largely ignore vertical sections; in part to remain as close as possible 
to Westfall’s text, in part to keep things as simple as possible.8 For the 
same reasons, I will ignore complexities arising from the fact that part of  
a building may exemplify a type without the building itself  exemplifying 
that type.

III.

The expression that is bound to cause most puzzlement is ‘natural 
symbol’. In the sense that is relevant here, a symbol is an object (or an 
action or an event) that stands for something valued within a community 
such as a deity or the nation.9 Two additional clarifications may make 
this definition a bit more helpful. Firstly, an object stands for something 
valued only if  it (i) has the capacity to call it to mind and (ii) in virtue 
of  that capacity, is the object of  attitudes normally directed towards the 



65

isparchitecture.com

valued thing (for example, reverence and respect). 
Secondly, what a symbol stands for usually is not 
itself  an object or at least not an object of  sensory 
experience: it can be a past event, a supernatural 
being, a non-perceptual property, a disposition, 
and so on. In the present case, it is (the purpose 
served by) an activity of  a very general kind such 
as venerating, celebrating and dwelling. What 
distinguishes natural from non-natural symbols 
is the way the connection is established between 
the symbol and what it stands for. In the case of  a 
non-natural symbol, the connection is established 
by a convention such as the convention that x 
is a monument for y or that x is the flag of  y. In 
such cases, the connection is arbitrary in the sense 
that a different convention would have resulted 
in a different connection. In the case of  a natural 
symbol such as an icon or a relic, the connection is 
not the result of  a convention, but of  something 

Figure 1: Westfall’s alphabet of building (p. 160): (1) regia (2) 
tholos (3) temple (4) dwelling (5) theatre (6) shop. Hatched 
areas represent solid forms. 5



AP . vol 1 . No 1 . 2014

66

De
 c

le
r

cq

outside our will such as striking resemblance and/or spatiotemporal 
contiguity (contact). Note that the difference lies in how the connection is 
established – by convention or not – not in how it is maintained. Even if  
the connection between a symbol and what it symbolizes was established 
by convention, a change of  convention may not result in a change of  
connection. For example, a country may adopt a new flag without the old 
flag thereby losing its symbolic significance. 

Evidently, this raises the question how the ‘natural’ connection is 
established between building plans of  a given type on the one hand and 
kinds of  activities on the other hand. Westfall does not explicitly address 
this question, but, as we will see, there is an answer that is compatible with 
what he says and which seems defensible. 

One may think that spatiotemporal contiguity can explain the symbolic 
connection between building plans and activities. After all, the symbolized 
activities (for example, dwelling) may take place in close proximity to the 
horizontal sections of  buildings. However, on the face of  it, this cannot be 
the whole explanation. The reason is that spatial contiguity can establish 
a connection only between items that have a location in space and time, 
for example, between a particular – locatable – building and a particular 
– dateable – activity that has taken place within it. But the connection 
that Westfall needs is one between building plans of  a given type and 
certain kinds of  activities. Because types and kinds are abstract objects, 
which do not have a location in space, they cannot stand in a relation of  
spatiotemporal contiguity to one another.

One might think that it is enough to have spatiotemporal contiguity 
between tokens and instances in order to have a connection between the 
corresponding types and kinds. In other words, if  activities of  a certain 
kind always take place in buildings of  a certain type, or vice versa, if  
buildings of  a certain type always accommodate activities of  a certain 
kind, then will this not be enough to establish a symbolic connection 
between the two? To be sure, this seems closer to the truth. It certainly 
is true that the actual use of  a plan may enhance its symbolic value. By 
having a certain history of  use, a plan may be better able to call to mind 
what is symbolized, for example, through recollection. However, it does 
not follow that such constant association can create symbolic value or that 
it creates it in every case. At least in our present case, it seems that other 
factors may be involved. In fact, in order to find out whether there are 
other factors, it may be worth investigating why certain types of  plan have 
come to be associated with certain kinds of  activity. For example, it may 
be worth asking why veneration often takes place in tholos-like forms. 



67

isparchitecture.com

An obvious answer is that this type is experienced 
as suitable or appropriate. This then suggests the 
following answer to our initial question: a building 
plan of  type x can be a natural symbol of  a kind 
of  activity y because x is (especially) appropriate 
for the accommodation of  y. Or, if  one prefers the 
interpretation under which purposes rather than 
activities are symbolized: a building plan of  type x 
can be a natural symbol of  purpose y because x is 
especially appropriate for the accommodation of  
an activity that serves y.

One may ask what is more important, whether 
a building plan is experienced as appropriate or 
whether it really is appropriate? I think the latter 
relation is more fundamental, and in any case, in 
greater need of  clarification (to be provided soon). 
Nonetheless, it is obvious that appropriateness alone 
does not suffice to confer symbolic significance 
on a building plan. The appropriateness has to be 
recognized if  it is to make the building plan stand for 
a particular purpose or a particular kind of  activity. 
In other words, only manifest appropriateness 
explains symbolic significance. Similarly, a piece 
of  clothing cannot function as a relic unless it is 
believed to have belonged to a particular person.

The idea that certain types of  plan are 
appropriate for certain kinds of  activity is one that 
occasionally shows up in the literature, for example, 
when it is said that variations on a plan and/or 
section occur in the history of  building types until 
the ‘appropriate’ form has been found. To quote 
just two examples:

In a healthy building culture, tradition and innovation 
are not contradictory but complementary concepts. In 
both cases, [a] building type is allowed to take on the 
most appropriate form (Davis 2006, p. 153; italics 
omitted).

A type represents the organizational structure of  a 
building in plan and section. A type evolves until it 



AP . vol 1 . No 1 . 2014

68

De
 c

le
r

cq

achieves its basic (i.e. its rational and logical) form… [T]he fact that airport 
terminals are everywhere in a state of  permanent reconstruction demonstrates the 
fact that shelter and function have not yet found a suitable type. (Krier 2007, 
p. 42) 

Although ‘suitable’ and ‘appropriate’ seem to be used here in their everyday 
meaning (of  ‘right’), it may be useful to spell out the idea a little bit more.

Echoing Nelson Goodman’s theory of  rightness (see, for example, 
Goodman & Elgin 1988, p. 158), one might say that a type of  plan is 
appropriate for an activity if  and only if  using it for the accommodation 
of  that activity works; in other words, if  using the type for that purpose 
helps to create a successful building, where ‘helps to create’ can be 
understood as designating a causal relation. Nothing much hinges here on 
a particular view of  what a successful building is, but it seems reasonable 
to demand – in addition to the Vitruvian criteria of  firmness, commodity, 
and delight – that a building be adaptable and in harmony with the existing 
environment.10 

If  using a type for the accommodation of  a particular activity works, 
then at least part of  the explanation must be that the use of  the type 
fits into a larger building culture which itself  works.11 As Goodman and 
Elgin say, “working tests fitting” (Ibid., p. 159). For example, architects 
and building users unfamiliar with a type of  plan may have no clue as to 
how a building based on it can accommodate the activity it is supposed 
to accommodate, and they may find it difficult to fit it into an existing 
context. In the building culture to which they belong, using the type may 
not work.12 When confronted with such a case, one should not jump to 
the conclusion that there is something wrong with either the type or the 
culture. One may just be confronted with a lack of  fit between the two. 
Nonetheless, the possibility of  a deficient building culture should be 
borne in mind, especially when criticisms of  the culture are widespread 
and persistent. This seems to be the case with our present building 
culture, whose shortcomings have been foregrounded by an ever-growing 
number of  critics, including architects (e.g., Bruce Allsop, Léon Krier, 
Paul Rudolph), engineers (e.g., Malcolm Millais), architectural critics (e.g., 
Peter Blake), architectural theorists (e.g., Brent C. Brolin, Howard Davis), 
architectural historians (e.g., David Watkin), philosophers (e.g., Roger 
Scruton), sociologists (e.g., Nathan Glazer), writers (e.g., Jane Jacobs, Tom 
Wolfe), and the Prince of  Wales (Charles).13 According to these critics, our 
present building culture is not very good at producing successful buildings 
in the sense that has just been made half-precise; in other words, it does 
not work very well.



69

isparchitecture.com

The notion of  a successful building, then, 
turns out to be essential to spelling out the relation 
of  appropriateness between building plan and 
activity (which relation is essential to explaining the 
symbolic significance of  certain types of  building 
plan). Of  course, this makes the relation partly 
normative. From a hermeneutic point of  view, this 
is a desirable consequence, since Westfall often 
calls his types ‘normative’. However, Westfall’s 
types are normative in a stronger sense: they are 
not just appropriate forms, they “provide the basic 
character of  what the plan of  the finished building 
must be if  it is to serve its purpose” (p. 161; my 
italics). Moreover, it seems that, in Westfall’s view, 
the only condition that can get one exempted from 
the requirement of  using the form for a particular 
activity is one in which it is (practically) impossible 
to use the form for that activity: “while the building 
types suggest how certain purposes ought to be 
accommodated by a building, circumstances do not 
always allow it to be so” (p. 161; italics in original). 
This may be overly strict. Nonetheless, it is worth 
noting that, according to Westfall, the requirement 
to use a certain building plan applies only to certain 
kinds of  building. For example, he claims that 
it does not apply to farms and libraries, and for 
reasons to be given soon, one may suppose that it 
does not apply to factories either. Moreover, the 
issue of  whether the types are normative in a strong 
or in a weak sense (i.e., whether they are mandatory 
or merely appropriate) has little bearing on the 
main issue, that is, how building plans can function 
as natural symbols of  certain purposes or activities. 
After all, if  a building plan is required in a given 
context – if  one “must” use it – then, a fortiori, it is 
appropriate to use it in that context.

The normativity imparted by the notion of  a 
successful building may seem to be incompatible 
with the claim that building plans are ‘natural’ 
symbols. However, the distinction between natural 
and non-natural symbols is not supposed to be an 



AP . vol 1 . No 1 . 2014

70

De
 c

le
r

cq

instance of  the metaphysically puzzling distinction between the descriptive 
and evaluative, causes and reasons, and so on. Rather, the distinction 
was supposed to be a metaphysically innocent one between items whose 
symbolic significance is based on a convention and items whose symbolic 
significance has a different basis. (By ‘symbolic significance’ I mean 
an item’s standing for something valued.) In sum, ‘natural’ has to be 
understood as ‘non-conventional’.

Even if  an item’s symbolic significance is not the result of  a 
convention, there is still a potential for cultural variability. To see this, 
consider that, according to the explanation given above, appropriateness 
is a matter of  working, and that working is a matter of  fitting within a 
culture of  building. As a result, instances of  the same type of  building 
plan may not stand for the same thing, or even anything, in every culture. 
It does not follow (from the fact that working is a matter of  “fitting within 
a culture of  building”) that there is no fact of  the matter as to what is 
a working building plan or even a working building culture. Compare: 
from the fact that different types of  medicine work for different kinds 
of  people, it does not follow that there are no absolute standards for 
therapeutic effectiveness. Moreover, although I have again departed from 
what Westfall suggests – like Jungian archetypes, his basic building types 
seem to have universal symbolic significance – our disagreement (again) 
has no implications for the main thesis.

As said, there is room for variability across cultures with respect to 
what a building plan stands for. However, within a culture, a building plan 
cannot stand for a completely heterogeneous collection of  activities, even 
if  those activities happen to be all instances of  the same kind of  action. 
In other words, there have to be constancies in the form the activities 
take (p. 158). Otherwise there cannot be a type of  plan that suits them 
all, and which can therefore serve as a symbol of  the kind. If  this line of  
reasoning is correct, it may explain why the activities of  production and 
transportation do not seem to have associated types of  building plan: there 
is just too much variation in the form that such activities can take. In other 
words, even if  there is a type suitable to railway stations (cf. Krier 2007, p. 
51), there is no type suitable to all of  the following: railway stations, bus 
stations, ferry piers, airports, rocket launchers, and so on. 

Of  course, this raises the question of  when certain activities are too 
heterogeneous to be appropriately accommodated and hence symbolized 
by a particular building type.14 It seems to me that this question is best 
answered a posteriori, by examining the building types that have actually 
been developed. To be sure, certain expectations are reasonable on a 



71

isparchitecture.com

priori grounds. For example, it seems reasonable 
to expect that activities which are defined in 
highly general terms and which, at the same time, 
are highly reliant on technology (for example, 
transportation, production, and research) will not 
find appropriate accommodation in a single type 
or even in a plurality of  types. In such cases, the 
appropriate form of  accommodation is likely to 
depend on the nature of  the technology and the 
stage of  its development. However, it seems that 
such a priori reflection on the relationship between 
building types and activities can easily lead one into 
paradoxes. Consider, for example, the apparently 
contradictory statements which the prominent art 
historian James S. Ackerman issued about the villa 
(on what seem to be a priori grounds). One the one 
hand, Ackerman writes:

This makes the villa unique: other architectural 
types—the palace, the place of  worship, the factory—
have changed in form and purpose as the role of  
the ruler, the character of  the liturgy, the nature 
of  manufacture have changed, frequently and often 
radically. But the villa has remained substantially the 
same because it fills a need that never alters, a need 
which, because it is not material but psychological and 
ideological, is not subject to the influences of  evolving 
societies and technologies. The villa accommodates a 
fantasy which is impervious to reality (Ackerman 
1990, p. 9). 

On the other hand, he writes:

There is hardly a moment in the history of  architecture 
when villas were less innovative than other architectural 
types… The villa is less fixed in form than most other 
architectural types because the requirements of  leisure 
lack clear definition (Ibid., p. 18). 

On the face of  it, this seems inconsistent. How can 
the villa, and the activity or “need” it is supposed to 
serve, simultaneously be more stable and less fixed 
compared to other building types?15 One may try to 
resolve this apparent contradiction by saying that 

In the case of a 
natural symbol 

such as an icon 
or a relic, the 

connection is not 
the result of a 
convention, but 

of something 
outside our will 
such as striking 

resemblance 
and/or 

spatiotemporal 
contiguity 

“

”



AP . vol 1 . No 1 . 2014

72

De
 c

le
r

cq

certain elements of  the villa remain unchanged while others are subject 
to change. But how could the villa be “unique” in this respect? Unless 
reference is made to particular elements, the statement that some of  the 
villa’s elements change, while others do not, borders on triviality. Of  
course, the only way one could reasonably make reference to particular 
elements is by investigating the matter empirically (that is, a posteriori). 
Eventually, this is what Ackerman does by distinguishing two types of  
villa that have survived from ancient times: the ‘compact-cubic’ type, 
which is regular (cubical) in shape, and the ‘open-extended’ type, which 
is irregularly shaped and more integrated with the natural environment 
(Ibid., pp. 18-26).

IV.

The hypothesis put forward in the previous section is that a building 
plan of  type x can be a natural symbol of  a kind of  activity y because x 
is (especially) appropriate for the accommodation of  y. Alternatively, a 
building plan of  type x can be a natural symbol of  purpose y because x 
is (especially) appropriate for the accommodation of  an activity serving 
y. In brief, appropriateness explains symbolic significance. However, it 
seems that the two factors can be regarded as mutually reinforcing, since 
the reverse—symbolic significance explains appropriateness—is also true. 
When a building plan comes to stand for a certain activity, it becomes all 
the more appropriate to use it for the accommodation of  that activity. This 
is of  course most clear in the case of  buildings accommodating ‘serious’ 
activities such as monuments, government buildings and churches. In such 
cases, it may be found awkward and even desecrating if  the building plan 
calls to mind another, less serious activity; similarly, an ordinary dwelling 
or shop may be found pompous or pretentious if  its plan calls to mind a 
more serious activity. Furthermore, a configuration of  buildings wearing 
their purposes ‘on their sleeves’ is more likely to constitute a meaningful 
whole. After all, we do not experience a configuration of  buildings as an 
abstract configuration of  shapes and colors any more than we experience 
an individual building in such a way. Our knowledge of  the actual purpose 
of  a building inevitably informs our perception of  it. If  that purpose does 
not correspond to its apparent purpose, then this can be easily picked up 
as an oddity in an individual case, which may even enhance the significance 
of  the whole. However, if  such incongruence occurs on a larger scale, 
then our attention will tend to drift away from the whole to the individual 
buildings making up the whole. What we are left with, then, is no longer an 
environment properly speaking, but a sequence of  buildings whose true 



73

isparchitecture.com

significance is to be deciphered.

Finally, if  a building plan’s symbolic significance 
is somehow local—a possibility mentioned in 
previous section—then that may make it again 
more appropriate because it may create the ‘sense 
of  place’ that is so much desired nowadays.16 

V.

This paper took its starting point in a claim 
attributed to Carroll William Westfall, namely: there 
are buildings belonging to disparate functional kinds 
whose plans are natural symbols of  the (purposes 
served by the) activities accommodated by those 
buildings. My aim was to give an interpretation to 
this claim that makes it plausible and yet not trivial. 
Moreover, the interpretation was supposed to 
remain close enough to Westfall’s text to be able to 
count as an interpretation. In order to demonstrate 
the plausibility of  the claim under the proposed 
interpretation, I explored the idea that certain types 
of  building plan are appropriate for certain kinds of  
activities. My claim was that such appropriateness can 
explain the natural (i.e. non-conventional) symbolism 
of  building plans, and can also be reinforced by it.

Now that an interpretation of  Westfall’s claim 
has been provided, it is possible to say something 
about its significance. To be sure, Westfall is not the 
first to ascribe symbolic value to building plans. For 
example, Rudolf  Wittkower (1998[1949]) already 
argued that the circular plan of  certain Renaissance 
churches symbolized divine attributes and even God 
himself. In Eastern traditions of  architecture, such 
symbolism seems more entrenched. For example, 
Indian temple and palace plans were often based on 
cosmological diagrams called ‘mandalas’. Similarly, 
in the Temple of  Heaven Complex in Beijing, 
circular and square plans were used to allow for the 
worship of  (a round) Heaven and (a square) Earth 
respectively (Steinhardt 2002). To some extent, 
Westfall’s is a generalization of  such claims, since it 

On the face of 
it, this seems 
inconsistent. 
How can the 

villa, and the 
activity or “need” 

it is supposed 
to serve, 

simultaneously 
be more stable 
and less fixed 

compared to 
other building 

types?

“

”



AP . vol 1 . No 1 . 2014

74

De
 c

le
r

cq

ascribes symbolic value to the plans of  buildings belonging to disparate 
functional kinds: religious structures, but also, for example, houses 
and shops. Moreover, under the interpretation given in this paper, the 
significance of  Westfall’s claim extends beyond the art-historical. After all, 
on this interpretation, the symbolic significance of  building plans does not 
depend on anything contentious. In particular, it can be reasonably doubted 
that the circle resembles God or Heaven, but it seems much harder to doubt 
that certain forms are appropriate for certain activities, let alone that such 
activities exist. In other words, the circular plan symbolized something for 
Renaissance architects (if  Wittkower’s interpretation is correct) and for 15th 
century Chinese people, but the plans Westfall has in mind can symbolize 
something for everyone capable of  recognizing their appropriateness, which 
basically means… everyone. In short, Westfall’s claim is not so much an 
interpretation of  a particular (historical or regional) architectural practice, as 
an interpretation of  architectural practice tout court. Whether it is a correct 
interpretation of  that practice has not been decided in this paper, but, 
hopefully, it now looks a little bit more credible. 

The previous paragraph may, however, invite an objection. In particular, 
Westfall’s claim may seem to be not just a generalization of  earlier claims 
about the symbolic significance of  floor plans; it may seem to be an over-
generalization. To understand why, it may be worth recalling two necessary 
conditions for being a symbol: (i) having the capacity to call to mind 
something valued (ii) in virtue of  that capacity, being the object of  attitudes 
(e.g., reverence, respect) normally directed towards the valued thing. Now 
it may be obvious how these two conditions can be met when what is 
symbolized is God or one of  his attributes. If  we are believers, such ‘things’ 
automatically command our respect. But what if  the thing symbolized is, 
as in (one interpretation of) Westfall’s theory, something banal like trading 
or dwelling? My inclination is to say that these, too, command our respect, 
although it may be less obvious precisely because they are so commonplace. 
Nonetheless, there are times when the importance we attach to them 
becomes manifest. Mircea Eliade, for example, reminds us of  the rites that 
accompany “the passing of  the domestic threshold” (Eliade 1959, p. 25) 
and the settling in a new house (Ibid., p. 57). In a similar vein, John Ruskin 
points to the “sanctity in a good man’s house which cannot be renewed in 
every tenement that rises on its ruin” (Ruskin 1990[1880], p. 179). Such 
remarks are difficult to understand if  dwelling is considered trivial and 
devoid of  intrinsic value. 

Acknowledgments



75

isparchitecture.com

Thanks to Paisley Livingston, Yu-Ping Luk, 
and especially Caroll William Westfall for helpful 
comments and suggestions.

References

Ackerman, J. S. (1990) The Villa: Form and Ideology of  
Country Houses. London: Thames & Hudson. 

Allsopp, B. (1974) Towards a Humane Architecture. 
London: Frederick Muller.

Blake, P. (1977) Form Follows Fiasco: Why Modern 
Architecture Hasn’t Worked. Boston: Little, Brown. 

Brolin, B. C. (1976) The Failure of  Modern Architecture. 
New York: Van Nostrand Reinhold Company.

Charles, Prince of  Wales (1989) A Vision of  Britain: 
A Personal View of  Architecture. New York: 
Doubleday. 

Davis, H. (2006) The Culture of  Building. Oxford: 
Oxford University Press.

De Clercq, R. (2011) ‘Modern Architecture and the 
Concept of  Harmony’, British Journal of  Aesthetics 
51, pp. 69-79.

Eliade, M. (1959) The Sacred and the Profane: The 
Nature of  Religion. Translated by W. L. Trask. New 
York: Harcourt, Brace & World.

Glazer, N. (2007) From a Cause to a Style: Modernist 
Architecture’s Encounter with the American City. 
Princeton: Princeton University Press. 

Goodman, N. & C. Z. Elgin (1988) Reconceptions in 
Philosophy and Other Arts and Sciences. Indianapolis, 
Indiana: Hackett Publishing Company. 

Hamlin, T., ed. (1952) Forms and Functions of  Twentieth-
Century Architecture. Volumes III & IV. New York: 
Columbia University Press. 

Jacobs, J. (1992) The Death and Life of  Great American 
Cities. New York: Vintage Books. 

Krier, L. (2007) Architecture: Choice or Fate? London: 
Papadakis Publishers.

the plans 
Westfall has 

in mind can 
symbolize 
something 

for everyone 
capable of 

recognizing their 
appropriateness, 

which basically 
means… 

everyone

“

”



AP . vol 1 . No 1 . 2014

76

De
 c

le
r

cq

Millais, M. (2009) Exploding the Myths of  Modern Architecture. London: 
Frances Lincoln.

Pevsner, N. (1976) A History of  Building Types. Princeton, New Jersey: 
Princeton University Press.

Rudolph, P. (2008) Writings on Architecture. New Haven, CT: Yale University 
Press. 

Quincy, Quatremère de (1999) The True, The Fictive, and The Real: The 
Historical Dictionary of  Architecture of  Quatremere de Quincy. Translated by 
Samir Younès. London: Papadakis. 

Rossi, A. (1982) The Architecture of  the City. Cambridge, Mass.: MIT Press.

Ruskin, J. (1990[1880]) The Seven Lamps of  Architecture. New York: Dover 
Publications.

Scruton, R. (1994) The Classical Vernacular: Architectural Principles in an Age 
of  Nihilism. New York: St Martin’s Press.

Steinhardt, N., ed. (2002) Chinese Architecture. New Haven: Yale University 
Press.

Thadani, D. A. (2010) The Language of  Towns and Cities: A Visual Dictionary. 
New York: Rizzoli. 

van Pelt, R. J. & C. W. Westfall (1991) Architectural Principles in the Age of  
Historicism. New Haven: Yale University Press.

Watkin, D. (2001) Morality and Architecture Revisited. London: John Murray.

Westfall, C. W. (unpublished) ‘The formal leg’, chapter 20 of  The Architecture 
of  American Liberty.

Wolfe, T. (2009[1981]) From Bauhaus to Our House. New York: Picador.

Wittkower, R. (1998[1949]) Architectural Principles in the Age of  Humanism. 
London: Academy Editions.

Endnotes

1. At least occasionally, I will allow myself  to deviate from Westfall’s 
terminology and metaphysics. For example, I will not always heed his 
fine distinction between ‘functions’ and ‘purposes’, and I will remain 
neutral on the ontological status of  types.

2. Henceforth, all references will be to this book (van Pelt & Westfall 
1991) unless indicated otherwise.

3. The term ‘building type’ is used ambiguously in the literature, as 
Quatremère de Quincy’s historical dictionary (de Quincy 1999) already 
makes clear. Although de Quincy did not make the different senses 



77

isparchitecture.com

much clearer (pace Rossi 1982, p. 40, who claims that de Quincy “gave a 
masterly definition of  type”), it seems possible to distinguish at least the 
following two senses in the current literature. On the one hand (see, for 
example, Hamlin 1952 and Pevsner 1976), ‘building type’ is used to denote 
the functional kind to which a building belongs: factory, monument, train 
station, church, hotel, town hall, concert hall, school, prison, museum, 
shop, and so on. On the other hand (see, for example, Davis 2006 and 
Krier 2007), ‘building type’ is used to denote the structural kind to which 
a building belongs, as determined by its plan and/or section: centralized 
versus linear, Greek-Cross versus Latin-Cross, single-story versus multi-
story, row versus detached, with or without courtyard or loggia, and so 
on. Westfall seems to use the term in the second sense, and so will I. Of  
course, it is possible two combine the two senses, for example, when one 
speaks of  the dumbbell tenement or the Danish row house as a type. 

4. More recently, Westfall writes that “[t]he presence of  the building idea 
type within the diagram and then the plan allows the building to signify, 
or better, to express, the activity it serves and therefore to connect it 
to the purpose that the institution or arrangements serves within the 
civil, religious, or cultural order” (unpublished, p. 12; my italics). This 
suggests that building types symbolize certain purposes by symbolizing 
(“expressing”) activities serving those purposes.

5. In personal communication, Westfall pointed to a minor inaccuracy in the 
diagrams included in van Pelt & Westfall 1991 (p. 160). The diagrams in 
this paper should better match his intentions.

6. According to Westfall (personal communication), these specifications 
“enter with the conventional configuration that the [plan] diagrams have 
taken within a time and a place and with attention to the actual functions 
that serve the purposes”.

7. By ‘recognizable’ is meant ‘recognizable in their instances’. In other words, 
it must be possible to sort particular buildings according to the types. For 
example, Bramante’s Tempietto in Rome can be classified as an instance 
of  the tholos type, which means that the type is recognizable.

8. Westfall does not ignore the vertical dimension. However, in his view, 
differences along that dimension (e.g., elevation) do not help to differentiate 
types of  building, but merely regional versions of  types (see, for example, 
pp. 162-7). Westfall does not ascribe natural symbolism to such versions, 
in part because he seems to equate ‘natural’ and ‘universal’. As will become 
clear, this is one respect in my reconstruction may differ from the original 
view.

9. “Signs point to particular meanings such as functions and the relative 



AP . vol 1 . No 1 . 2014

78

De
 c

le
r

cq

importance of  similar things, while symbols embody the larger purposes 
that enliven the aspiration to live the good life of  justice and nobility” 
(p. 156). 

10. For an analysis of  architectural harmony, see De Clercq 2011. 
However, harmony can be understood here in a wider sense, including 
environmental friendliness.

11. Building cultures are usefully analyzed in Davis 2006, where the term 
‘building culture’ refers to “the coordinated system of  knowledge, rules, 
procedures, and habits that surrounds the building process in a given 
place and time” (Davis 2006, p. 5). Members of  the culture include 
contractors, clients, architects, building users, bankers, and so on.

12. Brolin 1976 provides several examples illustrating how the acceptability 
of  a floor plan may depend on social customs and cultural values (e.g., 
pp. 42-3, 66-7, 99).

13. References are in the bibliography.

14. Here ‘building type’ is used in the structural or structural-cum-
functional sense. See footnote 3.

15. Ackerman acknowledges the tension by concluding that the villa 
“poses a cultural paradox” (p. 34).

16. The connection between architectural types and local character is 
stressed in Davis 2006 (several places) and in Thadani 2010 (p. 694).

RAFAEL DE CLERCQ responds

Question: If  we must admit to some cultural variation for the view of  natural kinds to 
hold plausibility, does then that recognition require us to accept that culture itself  is a natural 
kind to preserve the argument? In other words, if, as many think, culture is significantly 
dependent on conventions to function at a high level, then haven’t conventions crept in through 
the back door?

Answer: The argument in the article does not imply that building types 
are natural kinds; only that they are non-conventional symbols. So let me 
rephrase the question as follows: if  building plans symbolize by virtue of  
their appropriateness, and if  their appropriateness depends at least in part 
on culturally variable elements such as customs, then how can it be claimed 
that building plans are non-conventional symbols? My short answer is that, 
although the appropriateness of  building plans depends on the presence 
of  conventions, it is not itself  a conventional property. Rather, it is a causal 
property (cf. “one might say that a type of  plan is appropriate for an activity 



if  and only if  … using the type for that purpose helps 
to create a successful building, where ‘helps to create’ 
can be understood as designating a causal relation”). 
Perhaps the following example can help to understand 
how a property can depend on conventions without 
being itself  conventional. In Western societies, 
handshaking is among the conventions for greeting. 
As a result, refusing to shake hands may cause 
someone to feel offended. When this happens, the 
relation between the refusal and the feeling is causal, 
but it is also dependent on the existence of  a particular 
convention for greeting.

Q: What does an analysis of  architecture as captured by 
symbolism or type do or speak to, within the philosophical or 
architecture discourse? For example, the Westfall concept you argue 
for could be seen as a potential support for the traditionalists—
Leon Krier, Prince Charles, and their fellows—in their struggles 
with modernists. Do you agree and is this how these ideas apply?

A: I am inclined to think that the main argument in 
the text—plans symbolize activities by virtue of  being 
appropriate for these activities—is neutral between 
modernism and traditionalism. Nonetheless, it is a fact 
that traditionalists tend to be more concerned with 
building types, as can be gleaned from my references. 
Moreover, there is another argument in the text that 
may provide some support for traditionalism, namely, 
where it is said that “[w]hen a building plan comes 
to stand for a certain activity, it becomes all the more 
appropriate to use it for the accommodation of  that 
activity”. For various reasons, local building traditions 
tend to contain types with considerable symbolic 
significance, and this is a reason for drawing on them 
if  the quoted sentence is true.

Q: Kant’s distinction between free and dependent beauty 
bears a relationship to what you discuss here. For Kant, 
the beauty of  architecture was decisively dependent on its 
functioning as intended. Yet ultimately he opted for the idea that 
it wasn’t function per se, but instead a concept of  functioning 
that determined the harmonious play of  mental faculties that 
created an aesthetic experience of  architecture. This idea has led 



AP . vol 1 . No 1 . 2014

80

De
 c

le
r

cq

Richard Hill, for example, in his Designs and their Consequences, to argue that “We 
do not need actually to use a building in order to respond aesthetically to it: it will be enough 
that we see those aesthetic ideas that are incorporated into its design.” Could you respond to 
this more indirect, or perhaps once-removed sense of  utility as an alternative?

A: I certainly agree that we do not need to use a building in order to 
experience it aesthetically. But the question probably is how, or to what extent, 
a building’s aesthetic or architectural value is determined by its symbolizing a 
certain activity or purpose. It is difficult to answer this question in a general 
way. For example, in some cases, symbolizing a certain activity may be a 
function of  the building, in other cases, it may not be. Similarly, in some 
cases, the symbolized activity may be one that the building was designed 
to accommodate; in other cases, it may be one that the building ended up 
accommodating; and there are still other possibilities. The notion of  function 
enters here in different ways, and accordingly, there will be different answers 
to give for the different cases. In another paper (‘Reflections on a Sofa Bed: 
Functional Beauty and Looking Fit’, Journal of  Aesthetic Education 47:2 (2013)), 
I argue that not all functions are equally important from an aesthetic point 
of  view; for example, I argue that acquired functions are less important than 
original functions. If  this claim is correct, then it is has implications for how 
the above question is to be answered. But in any case, I think there is no doubt 
that symbolization of  activities is aesthetically relevant. This much is already 
implied by the sentence quoted in response to the second question, namely, 
“[w]hen a building plan comes to stand for a certain activity, it becomes all the 
more appropriate to use it for the accommodation of  that activity”.


