





































A new interpretative taxonomy 
for works of architecture

Rick Fox

I. Recent Interpretive Practice 

A core aim of  architectural interpretation 
is to elucidate, explain, make sense of, and help 
others understand the existing built-environment, 
proposed works of  architecture, architectural 
theories, and other elements of  architectural 
discourse. Indeed, there are numerous aspects of  
architectural works, theories and discourse about 
which we may render interpretations.  The field 
of  architectural interpretation broadly construed 
includes, inter alia, the literature of  philosophical 
aesthetics, the writings of  design professionals, 
architectural thinkers and educators, the views of  
architectural critics, and the voices of  the general 
public.

In this essay, I aim to introduce readers to a 
new philosophically grounded approach I have 
developed for organizing interpretive positions 
within architectural discourse.  The interpretive 
taxonomy I advocate relies on some philosophical 
insights from the American philosopher Michael 
Krausz. I will concentrate my efforts on presenting 
an overview of  the taxonomy, and focus on its 
importance to the philosophy of  architecture.  
To illustrate various elements of  the framework, 
I mention particular works, and discuss specific 
views, but the focus here is not on a detailed 
examination of  specific works, nor do I offer a 
robust rejection or endorsement of  any particular 
view. Before I get into the framework’s specifics, let 
me try to establish the breadth of  the problem, by 
offering three examples where interpretive practice 
would benefit from a more rigorous understanding 



AP . vol 1 . No 1 . 2014

98

fo
x

of  the space of  interpretation. 

Whether a new work of  architecture ought to “fit in” with extant 
works is an enduring problem in architectural theory and practice. 
Frequently, architectural works are criticized on the grounds that they do 
not belong. Many famous, and now respected, works have been criticized 
for this supposed failure.   

The contextualist argument relies on the claim that the existing context 
deserves our respect, and we ought to show deference to current norms 
and practices in our decision-making about how to, or even whether 
to, alter it. The argument, however, also makes a deeper supposition, 
namely, that the context is comprised of  self-evident facts and meanings. 
Yet, understanding what is “already there” is often more than simply a 
straight-forward documentation of  a few observable facts; it is a matter 
of  active engagement and interpretative practice. So, the background 
views of  the interpreter as to whether the context is “fixed” play a 
relevant role. The way we construe the boundaries of  a given context, and 
determine which aspects of  it are salient for our understanding is also a 
matter of  interpretative practice.    When we encounter a familiar work 
of  architecture we do not feel compelled to interpret it; we simply take it 
for granted. An alien or anomalous work that does not comport with our 
current understanding challenges us, and we are inclined to try and make 
sense of  it. Indeed, the views of  the interpreter may be affected in the 
process. The anomalous thing may prompt some of  us to recontextualize 
a few of  our existing beliefs as we attempt to make sense of  the world and 
the anomaly. Others prefer instead to reject the anomaly and adhere to 
their cherished certainties. Commonplace interpretive practices regarding 
context fail to account for its malleability, and the dialectic nature of  
interpretive practice. 

The rebuilding efforts on New York’s World Trade Center site 
following the events of  9/11 illustrate a second type of  interpretive 
problem. Significant disagreement exists in answer to the question, What 
does 9/11 represent? Answers that have been offered include, but are 
not limited to: 1) the unprovoked attack marks the turning point from 
the Cold War to the “War on Terror;” 2) it is a pointless act perpetrated 
by psychopaths who simply want to kill vast numbers of  innocent people 
as a demonstration of  their cruelty; 3) Islamic fundamentalists seek to 
end Western imperialism by any means possible—no matter how violent; 
and, 4) the attack is a misguided hatred of  global democracy and global 
freedom. For our purposes, the precise answer does not matter all that 
much, since it is not the real focus here. No matter what specific answers 



99

isparchitecture.com

are offered, the question itself  calls our attention to 
a broader philosophical concern, namely, is there 
a single right interpretation of  these events or 
not? Furthermore, it is clear that the meaning of  
the underlying events, no matter how we construe 
them, significantly affect our interpretation of  
the architectural work that is supposed to denote, 
memorialize, exemplify, or express them. In these 
sorts of  cases, our background knowledge, beliefs, 
and values play a significant role in the interpretive 
process. These background views often include 
beliefs about what role architecture ought to play, 
if  any, in responding to important social, political, 
and historical events.  

To further complicate attempts to interpret 
the meaning of  and best response to 9/11, the 

View of New York’s world trade center



AP . vol 1 . No 1 . 2014

100

fo
x

Byzantine array of  stakeholders has been the source of  a number of  
complicated political, economic, aesthetic, and interpretive disagreements 
over whose interests the work ought to advance, and whose stake ought to 
take precedence.  When the results of  the international design competition 
sponsored by the Lower Manhattan Development Corporation were 
unveiled in Fall 2002, the popular response was unprecedented.  Unlike 
most of  the other entries, Daniel Libeskind’s proposal resonated loudly 
with prominent architectural critics and the public. Ada Louise Huxtable 
emerged as an early advocate of  Libeskind’s proposal and remained an 
unwavering supporter of  the priorities expressed in his design.  Conversely, 
Herbert Muschamp, architecture critic for the New York Times, dismissed 
the rich symbolism employed in the design as jingoistic. John Silber, 
professor emeritus of  philosophy and law, and former president of  
Boston University, scathingly criticized the proposal calling it “an exotic 
and enticing jumble of  novelties” that expresses “hubris and a penchant 
for absurdity.” 1 Still other observers felt strongly that the only correct 
response to the tragedy was to rebuild exactly as before. How could 
such divergent interpretations all be, in some sense, correct or at least 
admissible?   

We take it for granted that all of  these critics are interpreting the same 
work of  architecture. This commonplace assumption, however, overlooks 
deeper philosophical questions about the nature of  the thing they are 
interpreting. For instance, it is a commonplace argument nowadays 
that complex things (such as a work of  architecture) license a plethora 
of  interpretations, and our contemporary society is prepared to regard 
so many things as indeterminate anyhow. Additionally, it is commonly 
accepted that in a pluralist society diverse and divergent interpretations 
ought to be encouraged, and tolerance ought to be embraced. Many argue 
that non-judgmental acceptance of  a plurality of  viewpoints is the ideal 
we should strive for in a democratic society. But, as the controversies 
surrounding the various interpretations of  the Freedom Tower design and 
the 9/11 Memorial illustrate, it seems counter-intuitive to think we are 
genuinely obligated to countenance every interpretation offered. Is there 
a non-arbitrary basis for admitting some interpretations while excluding 
others? If  so, how would this impact interpretive practice?

The question of  how to properly construe the relationship between 
the three-dimensional form of  an architectural work and its meaning, a 
staple of  debate within architectural discourse, illustrates a third type of  
interpretive problem.  A celebrated debate in 1989 between Leon Krier and 
Peter Eisenman on the theme of  “Reconstruction versus Deconstruction”2 



101

isparchitecture.com

forms a case in point. At the time, the stylistic 
skirmishes over historicism and the “next wave,” 
of  which this debate was a part, garnered a great 
deal of  attention, and pre-occupied many of  us in 
the architectural community.  It came as no surprise 
at the time that, stylistically, Krier and Eisenman 
were far apart with Eisenman claiming that history 
had few relevant precedents for contemporary 
conditions while Krier argued that all the necessary 
precedents already existed prior to modern times. 
The enduring gulf  between them is rooted in their 
divergent views about what a work of  architecture 
is, how we understand it aesthetically, the role 
architectural discourse is supposed to play within 
culture, and how best to interpret what a work of  
architecture means. The underlying philosophical 
differences between Eisenman, the paradigmatic 
spokesperson for autonomy and Krier, the 
traditionalist, could hardly have been greater. 

While the urgency of  this particular debate 
seems to have waned, the deep, unresolved 
disagreement about what properly constitutes a 
work of  architecture, and how we are to interpret 
or explain it remains worthy of  our philosophical 
attention. In my view, both historicism and 
deconstructivism are examples of  interpretive 
practices that rest on a few philosophical mistakes. 
Historicism is built on the premise that an object’s 
determinacy secures only one right interpretation 
of  it while deconstructivism, an example of  critical 
pluralism, holds an object’s indeterminacy licenses 
an open-ended plurality of  interpretations. (The 
critical pluralist argues that: 1) there can be more 
than one good way to construe the initial object; 2) 
divergent  interpretations are always admissible; 3) 
differing interpretive aims provide differing criteria 
as to what constitutes a valid interpretation;  and, 4) 
disagreements among qualified interpreters can be 
reasonable yet they may not be reconcilable. What 
makes critical pluralism  attractive, according to its 
advocates is the intuition that taken together several 



AP . vol 1 . No 1 . 2014

102

fo
x

judgments may be inconsistent, yet each judgment could be individually 
true. Since the identity of  the thing being interpreted is not stable, it is not 
susceptible to fixed  interpretation. Thus, critical pluralism  gives us a polite 
way to  have our strenuous disagreements, and robust truth at the same 
time.) These sorts of  claims: that what a work of  architecture is necessarily 
entail how many interpretations of  it we ought to hold admissible, while 
widely accepted, are mistaken; the entailment is not necessary. 

As I think these examples illustrate, what we have in each situation, 
ultimately, is a stand-off. We need a better way to talk about interpretive 
practices within our discipline to transcend their intractability. Architectural 
practitioners, theorists, critics, educators, and philosophers of  architecture 
need new interpretive tools.  

II. A Krauszian-style Taxonomy

In this enterprise of  developing new tools we need a philosophically 
grounded approach to architectural interpretation that: 1) promotes 
interpretive tolerance, whereby architecture may be interpreted in a 
number of  ways, without fostering unrestricted license; 2) serves as a 
superior framework to others prevalent in contemporary architectural 
interpretation, notably critical pluralism; and, 3) is not based on the false 
assumption that an object’s ontology necessarily entails the number of  
interpretations we find admissible. 

One candidate for introducing a new basis for architectural 
interpretation can be found in the two-tiered framework advocated by 
Michael Krausz. This framework differentiates the interpretation from the 
thing being interpreted, called the object of  interpretation. I prefer the term 
interpretandum, because this makes clear that what is being interpreted 
could be a real object, an abstract object, a sense perceptible phenomenon, 
a theory, or a social practice, but is not necessarily a material thing. One 
of  Krausz’s major contributions to the philosophy of  interpretation is his 
insight about the logical non-entailment between our ontological theory 
of  the interpretandum and the number of  interpretations we hold as 
admissible of  it. This thesis is called the detachability thesis. I offer the 
following formulation:

D: Our ideal interpretive stance with respect to the number of  admissible 
interpretations of  a given common object is logically detachable from specific 
ontological theories.

It is this mechanism that de-couples ontological theories from 
interpretive ideals. Each enterprise is logically independent of  the other; 



103

isparchitecture.com

neither logically entails the other. Thus, it is this 
thesis about the logical non-entailment between 
specific ontological theories of  interpretanda and 
the number of  interpretations we hold about them 
that sets up the overall framework of  possible 
interpretive positions that I advocate. Next, I offer 
overviews of  two ontologies useful for architectural 
interpretation, followed by a discussion of  the two 
interpretive ideals that comprise this taxonomy.

II.I: Realism/Constructivism 

One difficulty with the literature of  architectural 
theory and professional practice is that typically 
design professionals, architectural educators, and 
theorists do not have philosophically articulated 
ontological views. This makes it difficult to discern 
the full extent of  their interpretive position. 
Although I discuss various ontological views about 
what an architectural object is, for the purpose of  
articulating various interpretive positions, I do not 
endeavor to resolve in this essay whether there is a 
correct ontology for works of  architecture, and if  
so, what it ought to be. 

The gulf  between realists and constructivists 
derives, at least in part, from the distinction between 
object-as-such and object-as-represented. The 
realist may not be able to say precisely how things 
are, but insists that there is a way that they in fact 
are. The realist holds that realism can be defended 
even if  access to the way the world is cannot be 
(fully) obtained. Constructivists hold that this 
is a distinction without a difference, and that the 
object-as-such collapses into object-as-represented. 
Since the object-as-such does no interpretive work, 
then according to the constructivist, it ought to 
be dropped. For the constructivist, objects are 
never simply “given,” and no fact of  the matter 
exists that grounds either the interpretation or the 
thing being interpreted. So, it is in this way that all 
attempts to segregate objects-as-such from objects-

One of 
Krausz’s major 

contributions to 
the philosophy of 

interpretation is 
his insight about 

the logical 
non-entailment 

between our 
ontological 

theory of the 
interpretandum 

and the number of 
interpretations 

we hold as 
admissible of 
it.  This thesis 
is called the 

detachability 
thesis

“

”



AP . vol 1 . No 1 . 2014

104

fo
x

as-represented fail according to the constructivist. Any interpretive 
framework that presupposes that interpretanda are constituted as matters 
of  fact is already predisposed to realist ontology; frameworks denying 
this are predisposed to constructivism. Such presumptions would be 
detrimental to the larger project of  showing how an object of  interpretation 
is constituted within interpretive practices, and that the ontology of  an 
interpretandum does not necessitate the number of  interpretations we 
ought to find admissible. 

For the aesthetic realist, aesthetic properties are reified; there are real 
properties in the work and these ought to ground our judgments about 
it. Aesthetic realism does not demand that for any given artwork only a 
single aesthetic description is admissible; more than one description may 
be admitted. The realist view of  what constitutes a work of  architecture 
is that a work is necessarily a “brick and mortar” building possessing 
aesthetic properties, and standing before us. The claim that the difference 
between a building and an architectural work ought to be grounded in 
the possession of  aesthetic properties is well entrenched. When the 
British architectural historian Sir Nikolaus Pevsner famously remarked, “A 
bicycle shed is a building; Lincoln Cathedral is a piece of  architecture,”3 
he likely meant that aesthetic properties make the difference. The ontic 
locus of  aesthetic quality is in the work. By implication, then, aesthetic 
judgments are seen as stating truths that occur in virtue of  some mind-
independent state of  affairs. Realists hold that a substantial notion of  
truth is a valuable resource in ridding ourselves of  interpretations we find 
inadmissible. Preference is a measure of  how an interpretation comports 
with descriptive truth. Thus, under architectural realism elucidation aims 
at explicating actual or emergent properties in the work itself  – it is an 
object-centered ontological enterprise. 

The constructivist would find these claims unsupportable. According 
to the constructivist, a work of  art is primarily a mental object; its physical 
presence is primarily a device for transmitting meaning from one mind 
to another. Advocates hold that aesthetic perception is theory laden and 
thus we ought to drop any talk in aesthetics of  an un-interpreted world. 
The architectural constructivist holds that the design is the work of  
architecture, and a building is just a “brick and mortar” instantiation of  
that work. Under the constructivist’s approach a work of  architecture can 
be construed to exist in multiple media; a floor plan drawing, a perspective 
rendering, and a 3-D computer model could simultaneously constitute the 
same design, and hence the same work of  architecture. In a recent interview, 
the writer and architect Iman Ansari poses a question to Eisenman that 



105

isparchitecture.com

probes the difference between the [architectural] 
object and the idea of  the [architectural] object. 
Eisenman responds, “‘Real architecture’ only exists 
in drawings. The ‘real building’ exists outside the 
drawing. The difference here is that ‘architecture’ 
and ‘building’ are not the same.”4

It is important to note that this distinction 
between ‘architecture’ and ‘building’ is possible 
even where no actual brick and mortar construction 
exists that can be pointed to as the finished work. 
Thus, under architectural constructivism elucidation 
characteristically aims at clarity of  consciousness—
at epistemic rather than ontological clarity. It is an 
interpretation-centered epistemic enterprise.

In Section 3, I will expand on these ontologies 
as a way of  further illustrating the structure of  
this taxonomy. The next major component of  the 
framework is comprised of  the interpretive ideals 
of  singularism and multiplism, to which we now 
turn.

II.II: Singularism & Multiplism

Krausz distinguishes and explicates two 
interpretive ideals. One he calls singularism; the 
other he calls multiplism. Andreea Deciu Ritivoi, 
offers a helpful summary, “An ideal is a critical 
stance that identifies the range of  admissible 
interpretations of  a given object prior to the conduct 
of  interpretive inquiry.”5 As Krausz uses the term 
in his interpretive philosophy “idealization” is best 
construed as something like the “desiderata of  
interpretive practice.” Thus, an “ideal” refers to a 
value or belief  about interpretive practice that is 
worthy of  being optimized and promoted – it is a 
norm of  practice. (In private conversation Krausz 
has stated that he does not intend “ideally” to be 
read in any Platonic way.)

Singularism is the view that there is a one-to-
one relation between the thing being interpreted 
and an interpretation, and several admissible 

Singularism is 
the view that 

there is a 
one-to-one 

relation 
between the 

thing being 
interpreted 

and an 
interpretation

“

”



AP . vol 1 . No 1 . 2014

106

fo
x

interpretations are conjoinable into a single “unified” interpretation.6 I 
offer the following formulation of  the singularist thesis:

S: ANY given object of  interpretation necessarily answers to one and only one 
ideally admissible interpretation. 

The singularist holds that incompatible interpretations cannot be jointly 
defended, and incomplete knowledge about competing interpretations 
ought not act as a barrier to embracing the ideal of  a single comprehensive 
interpretation. For the singularist, under ideal conditions there just is one 
and only one right interpretation.7 Although the singularist advocates the 
pursuit of  a single right interpretation it should be noted, however, that 
the singularist need not advocate infallibilism. The possibility exists, even 
for the singularist, that any given interpretation could be replaced with a 
better one. In this formulation the singularist understands the admissibility 
of  a given interpretation in bivalent truth-functional terms.

For any given object, when distinct interpretations are present, the 
singularist would argue that the true ones are conjoinable into a single 
coherent interpretation. Logically, a conjunction is true just in case 
each of  its conjuncts is true. So given a number of  interpretations, 
the adherent of  singularism first determines which are true and which 
false, and then proceeds to work out how the true ones are conjoinable. 
Contemporary architectural interpreters, particularly those committed to 
critical pluralism, are unlikely to hold a strict view of  S. Yet, I hold that 
a thickened description approach, where a “unified” singular view of  the 
work emerges, qualifies as singularism. Later, I will argue in more detail 
why I think Zaha Hadid holds this view. 

Multiplism is the view that a one-many relation holds between the 
thing being interpreted and interpretations of  it. I offer the following 
formulation of  the multiplist thesis:

M: SOME objects of  interpretation MAY answer to more than one ideally 
admissible interpretation.

This thesis claims that multiple non-convergent interpretations are 
ideally admissible and ought not be conjoined into a single “unified” 
interpretation. The multiplist is committed to the view that sometimes 
there just is no such singularity, and holds that more than one interpretation 
could legitimately be held at the same time. Though multiplism recognizes 
that while several interpretations may be possible, not all of  them 
should be considered equally admissible; further it is not the case that all 
admissible interpretations are equally preferable. This allows the multiplist 
to embrace certain singularist cases, but not vice versa. The multiplist 
could acknowledge that there are cases where a one-to-one relation 



107

isparchitecture.com

between an interpretandum and an interpretation 
could occur without committing to the truth of  
S. In this formulation the multiplist understands 
admissibility in multivalent terms, in Krausz’s 
words as, “reasonableness, appropriateness, aptness 
or the like.”8 For the multiplist, affirming one 
interpretation as admissible does not necessarily 
exclude others as inadmissible. Even though 
multiplism does not adopt a bivalent mode of  
determining admissibility, principled distinctions 
among competing interpretations remain possible. 
Thus not every would-be interpretation is 
admissible. So, the architectural multiplist is keen 
to point out that rational inquiry about works of  
architecture does not mandate a bivalent “true-
false” assessment of  an interpretation.

III. Interpretive Positions

With this understanding of  ontological theories 
and interpretive ideals we are better equipped to 
understand the import of  the four main interpretive 
positions that a Krauszian-style taxonomy affords. 
Two of  these positions are orthodox; two are 
heterodox. The orthodox view is that constructivists 
promote a multiplicity of  interpretations while 
realists promote a single right interpretation. In 
my view, it is precisely their individual entrenched 
commitments to their respective orthodox views 
that deeply mired Eisenman and Krier during 
their rancorous 1989 debate. I offer Leon Krier, 
the American philosopher Michael Mitias, and the 
British philosopher Roger Scruton as characterizing 
the real-singularist position, and those of  Nelson 
Goodman and Peter Eisenman as characterizing 
the constructive-multiplist view. I would argue that 
in addition to being divided by their ontological 
commitments, these advocates embrace differing 
interpretive ideals that further divide them.

The detachability thesis, D, suggests that the 
heterodox interpretive positions of  constructive-

Multiplism is 
the view that 

a one-many 
relation holds 

between the 
thing being 

interpreted and 
interpretations 

of it

“

”



AP . vol 1 . No 1 . 2014

108

fo
x

singularism and realist-multiplism might be fruitful ones to hold. I offer 
two British architects, Zaha Hadid and Richard Rogers, both modernists, as 
examples of  constructive-singularism and realist-multiplism, respectively. 
While both of  them embrace the stylistic virtues of  architectural 
modernism, and on the surface this would seem to unite them, their 
ontological commitments and their differing stance about how many 
interpretations are ideally admissible creates an interpretive gulf  between 
them. Hadid’s commitment to modernism amounts to singularism, insofar 
as she holds that the proper role of  architecture is singularly fulfilled to 
the extent every project strives to be culture-altering and an opportunity 
to invade “new territories.”9 Rogers’ considerably more conciliatory view 
that architecture ought to be capable of  adapting to emerging social 
and technological circumstances, and be aesthetically resilient enough 
to endure alteration amounts to multiplism. While the views of  Hadid 
and Rogers characterize the heterodox interpretive positions, their views 
may not be exhaustive of  their respective positions. As with the orthodox 
positions, the views of  these interpreters ought to be seen as instances of  
the positions they occupy within the taxonomy, and not fully constitutive 
of  the entire position. 

Under Mitias’ realist-singular account, a work of  architecture emerges, 
that is it comes into being, in the aesthetic experience of  a building. Our 
experience of  the building as the physical structure of  an architectural 
work is the proper starting point for aesthetic inquiry. He rejects the claim 
that external factors such as background knowledge of  cultural practices 
and knowledge of  symbol systems are inherent to the work’s features qua 
object. According to Mitias, in his essay, “Expression in Architecture,” 
buildings possess and express their properties:

A building possesses its aesthetic properties. Those properties are not…added, or 
introduced, to the work from the outside, regardless of  the nature of  this external 
source. They originate from the building.10

Yet these aesthetic properties are not ready made realities, rather 
they exist as potentialities in the work. Aesthetic perception is needed to 
discover properties in the work “not given to ordinary perception.”11 In 
this way aesthetic properties are emergent in the aesthetic experience of  
a work. He writes:

This means that a building becomes art, i.e. acquires its aesthetic identity as 
a work of  art, only during an event of  aesthetic perception, and outside this 
perception its status is similar to the status of  ordinary objects.12

In a subsequent essay, “The Aesthetic Experience of  the Architectural 
Work,” he clarifies his view that aesthetic qualities are not feelings 



109

isparchitecture.com

evoked in the observer when aesthetic attention is 
brought to bear on the object, rather the quality is 
inherent in the object itself.13 His commitment to 
a realist ontology forces the conclusion that what 
distinguishes a building from a work of  architecture 
is the existence of  potential aesthetic properties in 
the latter but not in the former. Similar to Pevsner’s 
view, aesthetic properties are the ontic difference-
makers between ‘architecture’ and ‘building.’ 

Mitias embraces the view that there is only 
one set of  non-contradictory properties that can 
be said to emerge from a given artwork. The claim 
that aesthetic qualities inhere but not as ready-made 
realties, that is, as actual properties given to sense 
perception, but as potentialities in the work implies 
a certain singularism insofar as Mitias does not 
think that it is possible for several non-convergent 
aesthetic potentialities to emerge from the same 
building. The singularist is keen to point out that 
it cannot be the case that p & ~p are inherent in 
the same work. Mitias certainly does not want to 
countenance the idea that contradictory qualities 
could emerge from the same work for two different 
interpreters. If  two people hold contradictory 
views, then someone must be wrong. Taken 
together Mitias’ claims of  aesthetic realism are 
supposed to compel his conclusion that a single 
right interpretation of  a work is the goal of  inquiry.

In his essay, “How Buildings Mean,” Nelson 
Goodman claims that architecture is a building that 
symbolically functions, that is, a work of  architecture 
performs a referential function that a mere building 
does not. Although modernist works of  architecture 
typically do not denote anything outside themselves, 
at least not in the manner of  classical architecture, 
neither are they devoid of  meaning, contrary to the 
claims of  realists such as Krier. Under Goodman’s 
account the work qua object has features that allow 
it to point both outside itself  and to itself. A given 
work of  architecture has meaning in virtue of  its 

Hadid’s 
commitment 

to modernism 
amounts to 
singularism

“
”



AP . vol 1 . No 1 . 2014

110

fo
x

capacity to simultaneously embody, and make explicit reference to, the 
properties it possesses. Thus, a building’s meaning is rooted in what it 
exemplifies. One example he cites is Gerrit Reitveld’s Schroeder House 
(1924), Utrecht, where the work references its own structural features of  
columns, beams and walls.

One advantage of  Goodman’s constructivist account is that the 
difference between a building and a work of  architecture does not require 
any distinction between ordinary and aesthetic perception. Furthermore, 
his view that referential functioning, at least in part, ought to ground 
the divide that separates a building from a work of  architecture has an 
additional advantage for the constructivist, because this approach does not 
require an ontological resolution of  aesthetic properties. It only requires 
human understanding of  symbol systems. 

In Goodman’s view of  meaning, exemplification plays an important 
cognitive role with respect to works of  art, because it affords epistemic 
access to features we might not otherwise attend to. As this line of  
reasoning demonstrates, it is not incoherent to talk of  an internal locus for 
meaning, particularly for non-representational works, but focusing on the 
notion of  a locus tends to reify meaning and distract us from Goodman’s 
central point that the value of  a work of  art is its role in furthering our 
understanding, and that meaning is a correlate of  that understanding. 
As Catherine Elgin has remarked of  Goodman’s views, “Understanding 
works of  art is not a matter of  passive absorption, but of  active intellectual 
engagement with symbols whose syntactic and semantic features are often 
elusive.”14

His endorsement of  a multiplist ideal is explicit and unmistakable, “A 
work of  art typically means in varied and contrasting and shifting ways 
and is open to many equally good and enlightening interpretations.”15 The 
multiplicity of  interpretations Goodman is prepared to accept appears 
to be rooted in his commitment to constructivism. Goodman makes 
an important point about his view of  the nature of  architecture and 
interpretation. He writes “More than any other art, architecture makes 
us aware that interpretation cannot be so easily distinguished from the 
work.”16 His path to multiplism is evident insofar as what we think the 
work is due to our interpretations of  it – our interpretations thus construct 
the work.

Can a constructivist object to Mitias’ ontology yet concur with S? 
I hold that this is a reasonable result and I offer the Iraqi-born British 
Architect Zaha Hadid as an example of  a constructive-singularist. She 
is a constructivist due to her thorough-going reliance on the centrality 



111

isparchitecture.com

of  “constructed” symbol systems in her work. No 
determinate meaning exists to be discovered, so 
she claims. She is a staunch advocate of  aesthetic 
modernism and, in her view, this is the singular 
lens through which works of  architecture ought to 
be interpreted.17 Her unyielding commitment to a 
purified aesthetic modernism and her advocacy of  
the aesthetic implications of  modernity is clear:

We can no longer fulfill our obligations as architects if  
we carry on as cake decorators. Our role is far greater 
than that. We, the authors of  architecture, have to 
take on the task of  reinvestigating Modernity ... there 
is only one way and that is to go forward along the path 
paved by the experiments of  the early Modernists.18

In claiming that Hadid’s commitment to 
aesthetic modernism grounds her singularism, I 
am not making the larger claim that all adherents 
of  modernism are necessarily advocates of  
singularism. Nor am I making the claim that 
modernism is somehow the aesthetic equivalent 
of  singularism. In Hadid’s case her singularism is 
rooted in her commitment to modernism, yet this 
commitment in itself  does not entail that only 
singular interpretations emerge from modernity. I 
classify her as a singularist, largely because I think 
that a thickened description of  her work renders 
a single interpretation of  it the best approach. 
The thickened description argument for S relies 
on the distinction between “thin” concepts and 
“thick” concepts. Thick concepts are those where 
less variation is tolerated among instances. So, in 
the process of  building up numerous descriptions 
of  the thing under inquiry we narrow the range of  
admissible variations of  it, and thus come to see it 
as “unified.”

Stylistically, Hadid stands in stark contrast to the 
British philosopher Roger Scruton. In his book The 
Classical Vernacular Scruton argues that generations 
of  architects educated in a modernist outlook 
that rebuked traditional ornament, traditional 



AP . vol 1 . No 1 . 2014

112

fo
x

materials, and the orders of  classical Greek and Roman architecture have 
created aesthetically disappointing and nihilistic works that disregard 
the civic nature of  architecture.19 In short, he argues that classicism is 
more aesthetically correct than modernism. He is forthright in his disdain 
for modernism, “its language is uncouth, unredeemed by detail, utterly 
indifferent to its surroundings, or to the person who is obliged to pass 
by the building.”20 We need to see that his aesthetic commitment to 
classicism need not be tied to his interpretive commitment to singularism. 
A multiplist, for example, could hold the views expressed by Scruton that 
modernism is not aesthetically satisfying, but reject his claims about the 
sole “true” function of  architecture. Clearly Hadid and Scruton disagree 
about the aesthetics of  modernism. Even though both are singularists 
they are divided by their ontology.

In contrast to these two singularists, the British architect Richard 
Rogers stands as an example of  a realist-multiplist. He is a realist insofar 
as he is committed to the view that the thing being interpreted, the “brick 
and mortar” building, is constituted independent of  what interpreters 
think. Rogers holds that technology, scientific research, and an emphasis 
on function should serve as the basis for contemporary architecture. 
However, he also holds that architecture ought to be capable of  adapting 
to changing social and technological needs. He is a multiplist due to his 
deep reliance on an “open ended” architectural object that is in principle 
flexible enough to admit numerous interpretations. He writes:

Though a building must be complete at any one stage, it is our belief  that in order 
to allow for growth and change it should be functionally and therefore visually 
open-ended. This indeterminate form must offer legible architectural clues for the 
interpretation of  future users. The dichotomy between the complete and the open 
nature of  the building is a determinant of  the aesthetic language.21

Yet we need to be careful not to argue that an underlying indeterminacy 
of  the work is what necessitates his being a multiplist. Contrary to the 
deconstructivist position, this line of  reasoning would undermine M1, 
rather than bolster it. Instead, it is important to show that in principle a 
common object is countable yet it admits of  more than one interpretation. 
That Rogers holds this outcome to be both possible and desirable is 
evident in the following passage:

The building form, plan, section and elevation should be capable of  responding 
to changing needs. This free and changing performance will then become part 
of  the expression of  the architecture of  the building, the street and the city. 
Program, ideology and form will then play an integrated and legible role within 
a changing but ordered framework. The fewer the building constraints for the 



113

isparchitecture.com

users, the greater the success; the greater the success 
the more the need for revision and then programmatic 
indeterminance will become an expression of  the 
architecture.22 

Rogers points to an important set of  difficult 
cases, namely, those objects of  interpretation that 
change over time. These sorts of  cases present 
difficulty not because they threaten our interpretive 
ideals; both singularism and multiplism remain 
unaffected. The singularist, for example, would 
argue that the one-to-one relation still holds over 
time, because at any given time the single right 
interpretation is grounded in the state of  the object 
at that particular time. The multiplist would agree 
that an object’s changed status could, though not 
necessarily would, ground different interpretations 
than those rendered earlier, but not because 
multiplism is false. These cases are difficult, 
precisely because where something is physically 
altered, the question emerges, Do we have a new 
interpretanda or not? What would constitute a 
priori limits to alteration, such that the interpretanda 
remains “fixed,” is difficult to say but is the crux 
of  the interpretive problem facing contextualists 
discussed at the outset. 

Under a Krauszian-style interpretive account, 
the community of  informed interpreters would 
be relied upon to determine, on a case-by-case 
basis, whether or not alterations to a given work of  
architecture create a new object of  interpretation.23 
Furthermore, this same approach to resolving 
interpretive disputes about individual works would 
apply to the larger context within which a new work 
is to occur. Some works of  architecture change the 
context within which they occur, and some do not. 
This realization that a dialectical relation exists 
between our views about context, and our views 
about the identity of  interpretanda could help 
move the contextualist debate forward. 

As the World Trade Center site example 

Hadid and 
Scruton disagree 

about the 
aesthetics of 

modernism even 
though both are 

both singularists

“

”



AP . vol 1 . No 1 . 2014

114

fo
x

illustrates, the ways in which architecture ought to memorialize human 
events continues to be a significant source of  interpretive disagreement. 
This framework makes it clear that interpretive debates often occur 
on many levels simultaneously. And this may be a significant source 
of  their seeming intractability. When differences in belief  about the 
underlying events (what it is that is being memorialized) are conflated 
with how admissible interpretations elucidate architectural responses to 
it, disagreement is bound to occur. Additionally, the framework makes it 
evident that one could be a either a singularist or a multiplist with respect 
to an underlying event, and proposed responses to it. Whether a critic 
could be a singularist about an underlying event, and a multiplist in regards 
to proposes responses (or vice-versa) is a question this approach explicitly 
raises. 

This taxonomy also brings conceptual clarity to the Krier-Eisenman 
debate by providing an account of  how their respective positions are 
both grounded in specific, yet differing, ontological commitments and 
differing interpretive ideals. In spite of  their many differences, both 
Eisenman and Krier, in effect, endorse the view (though they would not 
have phrased it this way), that claims about what a work of  architecture 
is necessarily entail the number of  interpretations of  it we ought to hold. 
As this taxonomy makes clear, both Eisenman and Krier have overlooked 
a crucial philosophical point about the relationship between their own 
ontological views and their interpretive ideals.

IV. Responding to Objections

Several objections may be raised against this taxonomy. Here, I address 
a few. 

Objection 1: Many contemporary works of  architecture are indeterminate, 
and this undermines S.  This is the critical pluralist’s objection. I concur, 
that one feature many contemporary works have in common is their 
indeterminacy. The existence of  indeterminate works, however we 
construe them, is a separate issue from whether indeterminacy compels 
the adoption of  a specific interpretive ideal. Importantly, the success of  
both S and M depends necessarily on the notion of  commonality—the 
proper construal of  a common object. For differing interpretations of  
entities to meaningfully compete they must be about a common object 
of  interpretation; they must be about the same thing. As I have already 
argued here, the determinacy of  the object of  interpretation does not 
necessitate singularism, nor does indeterminacy necessitate multiplism.24 

Thus, indeterminacy does not threaten S1—it remains a defensible ideal. 



115

isparchitecture.com

Objection 2: The taxonomy is incomplete; there are 
more than two ontologies for works of  architecture. This 
objection is almost certainly true. There is at least 
a third major ontological theory – constructive-
realism – in this taxonomy that I discuss elsewhere, 
but given its philosophical complexities I chose 
to forego discussion of  it for two reasons: 1) 
the difference between internal and external 
constructive-realism makes it difficult to determine 
the relevance of  constructive-realism as an 
architectural ontology; and, 2) no one within the 
field of  architectural interpretation, as far as I know, 
has advanced anything to discuss. 

Objection 3: There are lots of  interpretations about 
architecture; not just four. True, insofar as we are 
talking about the number of  actual interpretations; 
they are inestimable. But an interpretation is not an 
interpretive position. The focus of  the taxonomy is 
to articulate a philosophically-grounded framework 
that fosters enhanced dialogue and comprehension 
of  interpretive positions and strategies. As I have 
argued, the number of  interpretive positions is 
not unlimited. My aim here is not to articulate and 
then adjudicate specific interpretations, but rather 
to enrich dialogue among disputants, build a few 
bridges, and get beyond the interpretive gridlock 
that has characterized architectural interpretation in 
recent decades. 

Objection 4: The detachability thesis, D, fails. For the 
sake of  brevity, I offer two reasons why D succeeds. 
First, each of  the four positions is coherently 
adoptable. There are no internal contradictions 
within any of  them. Second, the methodological 
issue of  which interpretive ideal we should adopt, S 
or M, can occur prior to interpreting any particular 
object, and it simply does not hinge on how we 
characterize the ontology of  the object we are 
interpreting; one issue is located at the level of  
practice, the other at the level of  ontology.

In conclusion, the interpretive framework 



AP . vol 1 . No 1 . 2014

116

fo
x

I propose has numerous advantages. First, the interpretive conduct 
it sanctions with respect to the number of  admissible interpretations 
does not entail an a priori commitment to a specific ontological theory, 
something architects do not possess as a matter of  professional training, 
and something that few architectural theorists articulate in any careful or 
systematic fashion. Second, where an interpreter is committed to a particular 
ontology with respect to works of  architecture, such commitment does 
not obligate them in advance to accept a specific number of  admissible 
interpretations. Being a realist, for example, does not compel an interpreter 
to embrace the view that a single right interpretation exists. Last, it is useful 
for assisting interpreters that may be unable or unwilling to overcome 
differences in their interpretive positions and reach agreement about 
the cultural value and meaning of  specific works. For all these reasons, I 
consider this interpretive framework to be a substantial improvement over 
what we have now.

ENDNOTES

1. John Silber, Architecture of  the Absurd: How “Genius” Disfigured a Practical Art. 
(New York: Quantuck Lane Press, 2007), 59.

2. Peter Eisenman and Leon Krier, “Peter Eisenman versus Leon Krier: My 
Ideology is Better Than Yours,” Architectural Design 59, 7-18.

3. Nikolaus Pevsner, An Outline of  European Architecture (Layton, UT: Gibbs Smith, 
2009), 10.

4. See “Q & A: Peter Eisenman”, in The Architect’s Newspaper, Vol7, Issue 5, May 
29, 2013. p. 22.

5. Andreaa Deciu Ritivoi, Introduction to Interpretation and its Objects, ed., Andreea 
Deciu Ritivoi (Amsterdam: Rodopi, 2003), 8.

6. Michael Krausz, Introduction to Is There a Single Right Interpretation?, ed., Michael 
Krausz (University Park: Pennsylvania State University Press, 2002), 1.

7. For proponents of  the singularist view see David Novitz, “Against Critical 
Pluralism,” in Is There a Single Right Interpretation? ed. Michael Krausz (University 
Park: Pennsylvania State University Press, 2002), 101-121; and, Brandon 
Cooke, “Critical pluralism Unmasked,” British Journal of  Aesthetics 42, 296-309. 

8. Krausz, Introduction, Is There a Single Right Interpretation? 2.

9. Zaha Hadid, “The Eighty-Nine Degrees,” in Theories and Manifestoes of  
Contemporary Architecture, 2nd Ed., ed., Charles Jencks and Karl Kropf  
(Hoboken: Wiley Academy, 2006), 280.

10. Michael H. Mitias, “Expression in Architecture,” in Philosophy and Architecture, 
ed., Michael H. Mitias (Amsterdam: Rodopi, 1994), 99.

11. Mitias, “Expression,” 99.



117

isparchitecture.com

12. Mitias, “Expression,” 100.

13. Michael H. Mitias, “The Aesthetic Experience of  the Architectural Work,” Journal 
of  Aesthetic Education 33, 64. 

14. Catherine Z. Elgin, “Relocating Aesthetics,” in The Philosophy of  Nelson Goodman: 
Nelson Goodman’s Philosophy of  Art, Vol 3, ed., Catherine Z. Elgin (New York: 
Garland Publishing, Inc., 1997), 171. 

15. Nelson Goodman, “How Buildings Mean,” in Reconceptions in Philosophy and Other 
Arts & Sciences, ed., Nelson Goodman and Catherine Z. Elgin (Indianapolis: 
Hackett Publishing, 1988), 44.

16. Goodman, “How Buildings Mean,” 45.

17. Hadid, “Eighty-Nine Degrees,” 280.

18. Hadid, “Eighty-Nine Degrees,” 280.

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

20. Scruton, Classical Vernacular, 16.

21. Richard Rogers, “Observations on Architecture,” in Theories and Manifestoes of  
Contemporary Architecture, 2nd Ed., ed., Charles Jencks and Karl Kropf  (Hoboken: 
Wiley Academy, 2006), 252.

22. Rogers, “Observations,” 253. 

23.What constitutes the community of  informed interpreters, and who qualifies for 
admission into it, is an important aspect of  the framework. For a more in-depth 
treatment see Chapter 3, Richard N. Fox, Interpreting Architecture: A Krauszian 
Approach. (Long Beach, CA: California State University Long Beach, 2009).

24. I first argued this point in Richard N. Fox, “Ambiguous Works of  Architecture: A 
Krauszian-style Interpretive Approach to the Gehry Residence,” (paper presented 
at the Pacific Division Meeting of  the Society for the Philosophic Study of  
Contemporary Visual Art, San Diego, California, April 2011).  In the paper I 
offered a critical extension of  the notion of  a common object of  interpretation 
and examined what we ought to look for in construing what the common object 
amounts to when we examine a specific work of  architecture. I argued that: (1) 
concerns about determinacy ought not compel us to adopt a particular interpretive 
ideal; and, (2) a work’s ambiguity does not hinder interpreter’s ability to agree on 
what constitutes the common object of  their interpretive disagreement.


