TYPAL AND TYPOLOGICAL REASONING:

A DIAGRAMMATIC PRACTICE OF ARCHITECTURE

Sam Jacoby
Architectural Association School of Architecture, UK

This is an Accepted Manuscript of an article published by Taylor & Francis in The

Journal of Architecture on 22.01.2016, available online:
https://www.tandfonline.com/doi/full/10.1080/13602365.2015.1116104

1
ABSTRACT

The twentieth century accounts of typology are often both historiographically

problematic and conceptually imprecise. They reinforce an understanding of typology
as mainly an interchangeable functional and graphic classification, and present Jean-
Nicolas-Louis Durand as a key figure of the discourse, despite him dealing with
buildings according to their genre and not their organisational and structural diagrams
of typology. In contrast, one can posit that all theories of type are foremost
epistemological and discursive arguments. Although not prescriptive in a formal sense,
they are concerned with a rational synthesis of form by thinking through conceptual
and diagrammatic organisation. This diagrammatic abstraction became instrumental
to architectural theory and history already in the eighteenth century, long before the
modern discourse on the diagram was consolidated in the 1990s.

While the architectural diagram is regularly explained as a generic and generative

description, it can be equally defined as a typological diagram specific to the
architectural discipline and its production of knowledge. Clarifying the concept of type
as emerging in parallel with ideas of abstraction and diagrammatic reasoning reveals
a richer set of connected problems deriving from architectural practice, pedagogy, and
disciplinary knowledge, which permits a different framing of the historical discourse.
This is explored by discussing its meaning for a distinction between typal and
typological reasoning, how this arises from a problem of history and theory, and how
the evolving typological discourse relates to the concepts of invention, disposition, and
style. Whereas historiography commonly recognises the French academics Antoine-
Chrysôthome Quatremère de Quincy and Durand, the often overlooked Gottfried
Semper and Julien-David Le Roy were central to a modern conception of architecture
that developed ideas of typal and typological abstraction through historicist processes
of cultural and diagrammatic reduction.

2
INTRODUCTION

The academic discourse on the architectural diagram consolidated in the 1990s. In

parts motivated by the technical possibilities of computational design and analysis, it
theorised design processes and problems of formal invention through the abstractions
of the ‘generative’ diagram. Despite emerging differences in definition, a common
ground to the debate was an agreement that a new diagrammatic practice had to
explain, analyse, organise, and generate beyond the limits of classical representation.1
This, so its claim to originality, liberated architecture from imitative repetition, gave it
autonomy (one not always already historicised), and produced something entire
different if not new.2 The explanations by key proponents of the diagram, such as Peter
Eisenman, Stan Allen or Robert Somol, were complicated by an avant-garde rejection
of representational ‘traditions’ while, at the same time, upholding that the diagram had
to somehow register architectural context, site, programme, history, and discourse.
Predictably, typology defined as a classificatory study of buildings with shared
functional and morphological traits, was seen as epitomising what the abstract diagram
is not: a restriction of generative and transformative reasoning by a pre-taxonomised
translation of conceptual and graphic thinking into materiality and architectural
objects.3

‘Diagrams underwrite all typological theories, as evidenced, for example, in the

catalogues of Durand’, writes Jeffrey Kipnis, suggesting that typology does not exist
without the graphic diagram.4 This statement—whose assumptions are shared by most
advocates of the generative diagram—reveals two flaws useful to the following
discussion. Not only can one equally assert that a typological problem underwrites all
architectural diagrams, but also Jean-Nicolas-Louis Durand did not deal with
typology—he abstracted buildings according to their genres (function) and was
unconcerned with typologies defined by comparable organisational and structural
diagrams of buildings. In addition, one can ask if there is more than a graphic diagram.
Yet Kipnis's mistake to employ typology in an interchangeable functional and graphic
sense is common to its twentieth-century use.

3
The Modern Movement deliberately reduced the nineteenth-century doctrine of type to
the functional classification of buildings to avoid its connotations of style, thereby
consciously eradicating distinctions between type and genre, but also type and
typology. When Giulio Carlo Argan ‘rediscovered’ the notion of type in his article ‘On
the Typology of Architecture’ of 1962, he revisited its first definition by Antoine-
Chrysostome Quatremère de Quincy, stating that a ‘typological series’—a series of
cases linked in their formal development—is determined by function and configuration,
and ‘has to be understood as the interior structure of a form or as a principle which
contains the possibility of infinite formal variation and further structural modifications of
the “type” itself’.5 Argan’s description of type as a form of knowledge internal to
architecture provides a rational explanation of the relationship between a historical
process and an architect’s individual design through a specific typological solution.
This explanation is autonomous from other disciplines, and complemented a return to
questions of historical and contextual continuity in post-war Italy. His interpretation
informed the ensuing typological discourse in Neorationalism, which, critical of
Modernist planning, saw the city and its elements not as a planning but design problem,
whereby a regulating typology linked to urban morphology to analyse context,
programme, and history could be mobilised.6 But, as Werner Oechslin argues, Argan
was also to blame for a widespread misconception of typology as iconology and its
decline to received forms.7 Although at first typology seemed to offer a sustainable
‘post’ Modernist design practice—an analytic architectural theory and urban science
as Aldo Rossi proposed in The Architecture of the City (1966)—its failure was by the
1980s widely accepted. Typology had become ‘a low level of theory’ providing little
more than fixed historical answers.8 Its use by the new discipline of urban design
conventionalised functional classification and graphic explanation of form.

Coinciding with a growing interest in design method, the spreading of the Neorationalist
theory of typology to Europe and America strengthened a focus on methodical
classification and design. This also applies to Alan Colquhoun, who, inspired by Tomas
Maldonado, in ‘Typology and Design Method’ (1967) is the first English-speaking
theorist to examine the notion of typology. Colquhoun contends that a final
configuration of form is never entirely an outcome of scientific deduction and involves
aesthetic intention. This intention, if it is more than intuition, has to acknowledge past
design solutions. Thus, typological models as repositories of existing formal solutions

4
and social meanings become necessary, but have to be adapted to a contemporary
context. Similarly, another important contributor to the English-speaking debate,
Anthony Vidler reinforced the narrative that typology is a problem of defining ideal type-
solutions, although he believes this was overcome by Neorationalism.9 In ‘The Third
Typology’ (1976), he sketches out three uses of typology. The first developing from an
imitation of nature to a scientific classification (from Marc-Antoine Laugier's primitive
hut to Durand’s collections of buildings), which also underlies the functional
classification of economic production by the Modern Movement and, third, a
Neorationalist understanding of typology as an analytic of the city through which the
‘unitary statement’ of form and function is transformed to an open possibility of
designing public architecture.

The discussed definitions of typology are largely consistent. The contemporary

accounts of typology, such as Vidler’s, permit the firm inclusion of Durand in the
discourse—which seems necessary to explain a functionalist and graphic definition of
typology—but are both historiographically problematic and conceptually imprecise.
The notion of ‘type’ formally entered architectural terminology only with the publication
of the third volume of the Encyclopédie méthodique: Architecture by Quatremère in
1825, where he presented it as an idea in contrast yet complementary to the model.
Thus, considerably later than either Durand’s Collection and Parallel of Buildings of
Every Genre, Ancient and Modern: Remarkable for Their Beauty, Their Grandeur, or
Their Singularity, All Drawn to the Same Scale (1799–1801) or the Précis of the
Lectures on Architecture given at the École Polytechnique (1802–5) that studied the
abstraction and derivation of buildings according to genres. Quatremère introduced the
term type to overcome two principles that he saw as preventing a modernisation of
architectural practice and knowledge: imitative representation and first origins. They
were characteristic for the Beaux-Art idea of the artistic model and stood for a
traditional theory of the arts and their teaching. As a modern concept, type replaced
previous categories of classification, such as character and genre. Yet throughout the
nineteenth century, its idea underwent continuous transformation in meaning, perhaps
due to the formal vagueness ascribed to it by Quatremère.

However, it can be posited that all theories of type are foremost epistemological and
discursive arguments. Although not prescriptive in a formal sense, they are concerned

5
with a rational synthesis of architectural (and urban) form by thinking through
conceptual and diagrammatic organisation. A concern with the discursive potential that
exists in the space from conception to formal realisation. It is a diagrammatic
abstraction that became instrumental to architectural theory and practice already in the
eighteenth century. This defines the architectural diagram not just as a generic and
generative description, but also as a typological diagram specific to the architectural
discipline and its production of knowledge. In this context, Durand’s graphic work
closely relates to a problem of type. The connection of type and diagram requires a
distinction between a conceptual (typal) and formal (typological) reasoning, consistent
with the one between idea and model found under the rubric ‘Type’ by Quatremère.
Through a typal reasoning, form acquires manifold historical, social, political, cultural,
and symbolic dimensions limited by but, importantly, also in excess of material reality.
The material and typological organisation of these social diagrams is in turn the
concern of spatial and graphic diagrams, which can also be termed typological
diagrams.

To explain the premise of a typal and typological reasoning and their relation to forms
of abstraction, one has to examine how this distinction and interrelation relates to a
separation of history and theory. How theories of type are framed by problems of
invention, disposition, and style—the first three principles of transformative
composition in rhetoric—through which arguments are conceived, structured, and
delivered, or considered as equally made up of conceptual, formal and social aspects.
To clarify the concept of type as emerging in parallel with ideas of abstraction and
diagrammatic reasoning reveals a richer set of connected problems that derive from
architectural practice, pedagogy, and disciplinary knowledge, and a different framing
of the historical discourse. Whereas historiography commonly recognises the French
academics Quatremère de Quincy (1755–1849) and his contemporary Durand (1760–
1834), the German architect Gottfried Semper (1803–1879) and the French
archaeologist Julien-David Le Roy (1724–1803) were central to a modern
understanding of architecture that developed from the typological discourse, but both
are overlooked in all key historiographical reviews.10 Similarly Le Roy’s formative
influence on Quatremère’s historical relativism, from which his theory of type derives,
has remained unnoticed.11 The three inaugural theories by Quatremère, Durand, and
Semper, despite different conceptions of type, share an understanding of form through

6
abstraction. Their historicist interpretations of form argue for a momentary autonomy
that arises from architecture’s constitution as an independent formal and artificial
language. In their theories and design studies, mimetic imitation is replaced by
memetic, conceptual, and symbolic abstraction as well as a diagrammatic reduction
that emphasises the interactions between type and diagram, abstraction and
translation, and idea and model. Therefore I will discuss what this means for an
understanding of type and typology, how this arises from a problem of history and
theory, and how the evolving typological discourse relates to the concepts of invention,
disposition, and style.

THE HISTORY AND THEORY OF ARCHITECTURE

A rising empiricism in the seventeenth and eighteenth century put architecture’s widely
held belief in classical authority into crisis. The need for modernisation led French
architecture to foster a modern canon tasked with consolidating past theory and
evolving practice. This relied in its studies of historical precedents and contemporary
architecture on empirical methodologies borrowed from emerging scientific
archaeology and dealt with problems of construction, often arising from new civic
structures and buildings. Recognising the waning of traditional explanations, Claude
Perrault therefore proclaims at the end of the seventeenth century:

Hence, neither imitation of nature, or reason, nor good sense in any way
constitutes the basis for the beauty people claim to see in proportion and in the
orderly disposition of the parts of a column; indeed, it is impossible to find any
source other than custom for the pleasure they impart.12

Perrault effectively declares an end to prevailing cosmological rationality and compels

a historicist relativism, however, one that can explicate the relevance of historical
precedents to current practice. This conception of conventional and relative
architectural styles required a new historiography. Thus by the late eighteenth century,
the claim of classical authority to universality was forever destabilised by historicism and
a recognition of individual expression and cultural diversity.

7
Critical for the historicist reassessment of architecture was Le Roy, who published The
Ruins of the Most Beautiful Monuments of Greece in 1758. In the book, he was the
first to introduce the concept of ‘history’ to architecture by distinguishing it from
architecture’s theory. This clarified the difference between a theoretical system of
principles, which Le Roy divides into three classes according to general and common
architectural ideas, and introduced history as the framework through which a
development of these principles can be understood.13 The first class contained
universal principles related to practical problems of construction and utility, the second
comprised of principles of perception and aesthetic judgement, and the third referred
to principles depending on climate, available building materials, and customs. These
geographically and culturally specific factors are therefore only selectively accepted,
nevertheless account for the variety of styles and formal differences. History, so Le
Roy, registers the development of architecture and its ‘primitive ideas’ in a series of
connected positivist and individual transformations. This relativises the problem of
origins, as higher artistic achievement of some people and cultures over others is
measured by a qualitative change and is not a simple question of chronology.

Le Roy analysed formal development through taxonomic comparison, in order to

determine relative stylistic periods and artistic achievements. He thereby noticed that
stylistic changes depended throughout antiquity on socio-cultural, geographic, and
climatic contexts and described a progressive yet nonlinear process of historical
contingency and exchange.14 But the irreversible conflict between representation and
reason after empiricism required a distinction between history and theory. When
history effects an architectural object, with its historicalness characterised by the
different contexts it registers, it also contextualises and effects the principles of
architecture, its theory. This makes a separation of history and theory necessary, and
arising from a tension between them, architecture exists then simultaneously as a
general (theoretical) and specific (historical) object that belongs both to the past and
present. As Le Roy explains in the second edition of The Ruins: ‘It is these differences,
these affinities, these successive transitions from one perfection to another that we
intend to demonstrate in the present essay.’15 To synthesise a metaphysical general
and a formal specific, means to read the architectural object as a historical object that
is judged by its presence (whether in the past or present), but also as belonging to a

8
continuous development of form. This on-going transformation is limited by ‘primitive
original ideas’, which due to their persistence throughout history can be deemed
ahistorical and as providing common criteria to the abstraction of a series of buildings
deriving from them. Formal development is effectively seen as occurring along a
typological line of development, with all instances sharing a comparable diagrammatic
trait. Consequently, comparative diagrams serve typological analysis and a judgement
of individual form against a theoretical possibility of form. They offer a simultaneous
theoretical and historical analysis of form, and suggest a separation and synthesis of
the knowledge that typal and typological reasoning make available.

The Ruins was significantly revised for its second edition in 1770, incorporating an
advanced argument of architectural history developed by Le Roy in the treatises History
of the Disposition and Different Forms That the Christians Gave to Their Temples since
the Reign of Constantine the Great to Our Own Day (1764) and Observations on the
Buildings of Ancient Peoples (1767). His idea of analysis is particularly apparent in the
History that summarises the evolution of churches in a comparative plate through the
juxtaposition of their plans and sections (Fig. 1). Evocative of a Linnaean taxonomy, the
matrix provides the arguments later adopted by all claims for an evolution of architectural
form: a process of methodical reduction and a diagrammatic explanation that relies on
comparison. While a graphic comparison of scaled plans itself is unoriginal, Le Roy’s
use differs from earlier instances, as his interest is not size, stylistic detail, or chronology,
but formal relations that describe a sequence of transformation and permit their
judgement.16 Published to demonstrate the superiority of Jacques-Germain Soufflot’s
design for Sainte-Geneviève over comparable designs, the church is depicted in the
centre of the plate as the synthesis of three developments, whose typologies are: the
cross-shaped plan, parallel rows of freestanding columns in the basilica, and the dome.
Explaining the importance of graphic abstraction to the representation and analysis of
architecture, Le Roy states: ‘A figure, even a small one, will better transmit an
understanding of a building and will more promptly communicate its disposition than the
most thorough verbal description.’17

9
‘Plan des églises les plus remarquables bâties depuis l’an 326 jusqu’en 1764’ [Plans of the Most
Remarkable Churches Built from 326 to 1764] by Jean-François de Neufforge, in Julien-David Le Roy,
Histoire de la disposition et des formes différentes que les Chrétiens ont données à leurs temples depuis
le règne de Constantin le Grand jusqu’à nous (Paris: Desaint & Saillant, 1764)

Examining the historical changes of the temple type in The Ruins (1770), Le Roy
introduces yet another important diagrammatic plate (Fig. 2). Organised into three
columns, it compares the sequential transformation of Egyptian and Phoenician, Greek
and Roman, and Christian huts into temples with increasing size and detail. The
diagram is accompanied by an extensive text explaining the relation of each instance
to its precedent in the imagined line of development that is depicted. Despite its
chronological appearance, which implies a linear development, the plate compresses
different historical developments into one comparative matrix, regardless of
chronology. Visible differences between instances manifest the contextual responses
through which history inflects the theoretical form of architecture.18

10
‘Plate 1’ by Michelinot after Le Moine showing the parallel formal development of the temple, in Julien-
David Le Roy, Les Ruines des plus beaux monuments de la Grèce: considérées du côté de l'histoire et
du côté de l'architecture, 2nd edn (Paris: Delatour, 1770)

Le Roy’s work asserts a productive relationship between architectural form and

historicity, proposing that form is historically specific and part of a larger and enduring
theoretical discourse. However, his interest is not to resurrect the past, but to didactically
use formal abstraction and historical knowledge to explain contemporary disciplinary
enquiry. To demonstrate how invention and disposition are closely related in the
development of form and can be methodically analysed and described. His thesis of
history prepares a modern reasoning that Michel Foucault characterised as ‘the
emergence of history as both knowledge and the mode of being of empiricity’.19

11
INVENTION

Quatremère in his winning submission for the Prix Caylus of 1785 closely follows Le
Roy’s position on development. Similar to Le Roy, who devised the theme and judged
the competition, he claims in the Mémoire on the Question: What Was the State of
Egyptian Architecture and What Do the Greeks Seem to Have Borrowed from It? that
the development of the tripartite origins of architecture in the cave, hut, and tent was
connected, despite emerging sequentially and in parallel in different cultures. After
editing the Mémoire for publication under a new title, On Egyptian Architecture,
Considered with Respect to Its Origin, Its Principles and Its Taste and Compared in
the Same Terms with Greek Architecture (1803), Quatremère radically overturned his
earlier conclusions. He now proposes common but multiple origins within different
cultures without any developmental connections. Unlike Le Roy’s positivist
development of typologies, it suggests an organic system of types. Architecture had
conceptually transformed from a natural to an artificial language with many filiations.
The fundamental revision emphasised a process of intellection through socialisation
and enculturation, which defined architecture and its knowledge production as a socio-
cultural appropriation unique to a society.20

Quatremère’s changing understanding of origins coincides with his appointment as

editor of the first French architectural dictionary, the Encyclopédie méthodique:
Architecture (1788–1825), in 1787, which required him to integrate architecture within
a new classification of knowledge that placed it amongst the fine arts.21 This meant
that he had to respond to the prevalent discourse of imitation, the conventional
framework to discuss artistic invention. To this debate on imitation, Johann Joachim
Winckelmann’s modern art history—influenced by Le Roy’s contextual reading of
artworks and their taxonomic comparison—was seminal in challenging imitative
principles, as he judged creativity as an abstraction of nature and not in classical terms
as a skilfulness to represent nature. Despite these precedents, Quatremère faced a
difficult double task. He had to establish architecture as an imitative fine art and
deconstruct its means of imitation.

Through the art historical discourse, he understood architectural imitation in relation to

the ideas of the ideal, resemblance, pleasure, convention, and invention, eventually

12
arguing against the still definitive principle of origins. He concluded this in An Essay on
the Nature, the End, and the Means of Imitation in the Fine Arts of 1823, in which he
defines imitation and invention as an intellectual abstraction requiring a re-composition
of reality through a vision or artefact that is socio-culturally specific and conceptually
ahistorical—showing the influence by Le Roy. It defines the arts as a form of typal
reasoning, with their production less a formal than a cultural abstraction that, although
limited by the practical and technical means available to each art, is principally only
constrained in its representation by social utility. While art is bound to a permanent
social contract, the production of artefacts is continuously transformed by changes in
technology, new materials, and cultural contexts. Developing Le Roy and
Winckelmann’s art historical theses, Quatremère considers imitation accordingly not
aesthetically but:

[…] abstractedly, that is, under a general and theoretical, and not a limited and
practical point of view, the words I may employ should be understood only as in
a sense related to the nature of an abstract theory, that is, one which
generalizes ideas.22

This non-mimetic, abstract quality was exemplified by artificial languages, especially

rhetoric. As architecture lacks a natural ability to imitate nature, its representations
require abstraction, which, so Quatremère posits, make the principles of transformative
composition in rhetoric available to architecture. Thus, architecture equally establishes
an artificial language. Representing a paradigm, pattern, and standard, type provides
to this language the important capacity to name, define, and communicate the
otherwise unknown—an abstract theory. In this sense of generalising abstraction can
Quatremère’s otherwise confusing use of ‘imitative’ be understood, when stating that
this communication is achieved by an imitative resemblance, in which an abstract idea
is translated into an engaging artefact. An artefact that due to the limitations of
architectural representation is always partial (in a naturalistic sense) and ‘produced
with and by means of elements distinct from the elements of that object’, indicating that
the realisation of an artefact is a ‘fictitious’ interpretation that simultaneously refers to
and differs from the object it represents.23 In this process, a generalisation through
abstraction becomes translatable into a generating type or conceptual idea, which has
the ability to obtain a knowledge unattainable to literal representation and

13
resemblance. Therefore, the incompleteness of representation is desirable as it
necessitates abstraction, and a precondition to decoding a typal idea in a typological
model. With the work of art principally unconstrained in its possible form, and imitative
resemblance signifying the abstraction of an ‘original type’, what becomes operative is
‘the principle of an abstract existence, of a nature very far removed from the principle
of identity’.24

Developing in the Essay on Imitation a system of abstract types against which material
objects are judged, Quatremère applies this conceptual diagram to architecture in his
dictionary entry ‘Type’ of 1825. The synonymity of the notions ‘image’ and ‘idea’ (and
‘ideal’ as an adjective of idea) is, as he points out, apparent from its etymological roots,
with idea deriving from the Greek eidos and eidolon that denote respectively a
conceptual type or Platonic Form and a physical apparition.25 Thus in ‘Type’ he
famously states:

The word type presents less the image of the thing to copy or imitate completely,
than the idea of an element which must itself serve as a rule for the model. […]
The model, understood in the sense of practical execution, is an object that
should be repeated as it is; contrariwise, the type is an object after which each
artist can conceive works that bear no resemblance to each other. All is precise
and given when it comes to the model, while all is more or less vague when it
comes to the type.26

While models have apparent rules, type represents a non-prescriptive ‘idea’, ‘motif’,
and ‘intention’. Type organises while the model structures. And typological models
serve a formal translation of speculative and non-material typal ideas. Contemplating
the closely related problem of invention, Quatremère concludes: ‘Everything must have
an antecedent; nothing whatsoever comes from nothing, and this cannot but apply to
all human inventions.’27 These ‘pre-existing seeds’ are found in formal precedents and
elementary principles, which as moral (intellectual) inventions are always obligated to
sentiment and taste. Accordingly, type is ‘like a sort of nucleus around which are
assembled, and with which are consequently coordinated, all the developments and
the variations of form to which the object was susceptible’.28 Although Quatremère

14
never explicitly explains how this formal variation is to be derived in practice, this is a
problem that, to some extend, was developed by Durand before him.

DISPOSITION

Quatremère’s typal reasoning in which conceptual abstraction is the basis to ‘invent’ a

disciplinary diagram of knowledge is complemented by Durand’s preceding
examination of architectural disposition that suggests a form of typological reasoning.
He describes architecture in functionalist terms as the formal disposing of parts, and
his comparison of abstracted historical forms made their reduction to formal diagrams
a means of analysis and design.

Accompanying his lectures at the École Polytechnique, Durand published the

Collection and Parallel (1799–1801), which in some editions included the Essay on
the General History of Architecture by Jacques-Guillaume Legrand.29 A student of Le
Roy, Durand was influenced in his abstraction of function to graphic diagrams by the
analysis of historical form through typological comparison. But Durand owed
methodologically more to a comparative method of classification by the zoologist
Georges Cuvier. This allegiance was in no uncertain terms asserted in Legrand’s
essay, claiming that through structural and formal analysis a ‘natural history of
architecture might be created’.30 Corresponding to Cuvier’s comparative taxonomies
in which history was reduced to formal and functional descriptions that lend themselves
to scientific analysis, and through which physiognomic development could be
reconstructed and explained, the ambition of the Collection and Parallel was to equally
employ history scientifically in architecture. Promising objectivity, technical drawing
rather than perspectival renderings provided the means to analyse formal
development. With history in the natural sciences defined as rational, as directly linked
to verifiable structural development, Durand considered the effects of style and
character on buildings as secondary and mere cultural phenomena. In their place,
structural relations and, implicitly, formal complexity became a material verification of
historical progress.

15
Durand developed the ideas of the Collection and Parallel in his Précis (1802-05) by
devising a design method simple to follow and instruct. As the Précis declares, its
method applies to the design of any building. Durand’s architectural course at the École
Polytechnique is therefore described as ‘the pursuit of certain ideas that are few in
number but general in application, and from which all the particular ideas would
necessarily derive’, outlining a ‘safe and rapid way to compose and execute buildings
of all kinds, in all places, and at all times’.31 Despite Durand understanding these
generative ideas in terms of function and differentiation of structure and not through
structural comparison, an affinity to the typological problem is evident. Yet Durand
justifies his generic method of design through common problems of utility. To achieve
utility, architecture has to be fit for purpose and maintain economy of means. According
to Durand, fitness derives from solidity (the right use of materials), salubrity (the right
choice of site and building exposure), and commodity (the right disposition of the
building), while economy relies on symmetry, regularity, and simplicity.

Durand’s design method relies on planar dispositions, with a horizontal plan informing
its vertical section. The disposition of a building and its elements develops from regular
grids and axes—a grid of parallel interaxis determined by the efficient structural
distance of two columns, according to which the building elements are distributed.
Subdividing the initial grid and omitting, adding, or offsetting one axis differentiates the
structural elements of a building. This creates unlimited part-to-part and part-to-whole
combinations of building elements and results in a mutation of the building parti, as
‘Plate 20’ demonstrates (Fig. 3). Despite the procedural nature of design, fitness of the
composition, so Durand, is also determined by the contextual requirements of ‘places,
persons, sites, costs, and so on’.32

16
‘Plate 20, Ensemble d’édifices, résultants de divisions du carré, du parallélogramme et de leurs
combinaisons avec le cercle’ [Building Ensembles, Resulting from the Divisions of the Square, the
Parallelogram and Their Combinations with the Circle], in Jean-Nicolas-Louis Durand, Précis des leçons
d’architecture données à l’École Royale Polytechnique, vol 1 (Paris: the author, 1802)

According to the method, once a plan is derived, the sections can be developed
through similar vertical combinations and, subsequently, plan and sections determine
the elevations. This sequence of design also exposes a basic problem of the method.
As the orthographic drawings of ‘Plate 21’ reveal, which shows supposedly a method
applicable to all architectural disposition, Durand’s plan-based process cannot logically
justify elevational drawings (Fig. 4). In order to determine these, conventions on scale
and mass are needed. The elevations and by implication the parti itself cannot be
produced by the proposed design method and rely on motivating precedent, as a dome
is unexplained by transformations of a planar grid.33

17
‘Plate 21, Marche à suivre dans la composition d’un projet quelconque’ [Procedure to Be Followed in
the Composition of Any Project], in Jean-Nicolas-Louis Durand, Précis des leçons d’architecture
données à l’École Royale Polytechnique, vol 1 (Paris: the author, 1813 edition)

What is therefore apparent is that the design method of the Précis depends on a
differentiation of precedents, which Durand admits to at the very end. He explains in a
later addition to the Précis, the Graphic Portion, that there is a difference between
learning to compose and composition proper. When learning, a didactic analysis of
part-to-part relations and eventually of the parts to the whole is important, however,
‘when you come to compose yourself, you must begin with the whole, proceed to the
parts, and finish with the details’.34 The design method is consequently conceived as
didactic and less a method of design than analysis, revealing Durand’s interest in a
didactic architectural project.

By breaking down the general idea of the architecture into special ideas, and
those into particular ideas, in the graphic portion we have broken down the
general ideas of buildings into those of their parts, and these in turn into those
of their primary elements; then, by working back from the elements to the
ensemble of the buildings—that is to say, by analyzing them—we have
succeeded in forming a precise idea of them, just as we had first succeeded in

18
 forming a precise idea of architecture itself by analyzing the general idea
expressed by that word.35

By abstracting precedents to operative diagrams, Durand is able to deploy received

form as contingent and not normative to design. His abstraction of architectural form
is, however, premised on an assumption of progression and limited to a taxonomy of
known work, as collated in the Collection and Parallel. Borrowing from the sciences,
Durand sees architectural knowledge as arising from a diagrammatic and generative
understanding of relationships. This gives architectural disposition a formal freedom
that, so Durand, is limited by society’s need for utility—defining thereby utility as a
social agenda. It is this generative understanding of form and functionalist conception
of the social, combined with an attempt to find a formal grammar capable of responding
to the exigencies of modern society, through which Durand anticipates the programme
and failure of the Modern Movement. The adoption of his design method by later
discourses of typology, convey the instrumentality and limitation that his graphic
diagrams bring to the discipline.

STYLE

Quatremère’s theories of invention and Durand’s exploration of disposition represent

first forms of typal and typological abstraction that make use of conceptual and graphic
diagrams. As complementary forms of reasoning, they became synthesised in
Semper’s work. Interested in the relationship between conception and materiality, he
examined how an abstract type is continuously realised in the material and technical
transformation of art-forms. Semper became familiar with Durand’s didactic teaching
during his studies in the 1820s in Paris, and witnessed a heated debate on polychromy,
which was first prompted by Quatremère’s The Olympian Jupiter of 1814. Inspired by
the problem of polychromy, Semper concluded his own archaeological studies in the
Preliminary Remarks on Polychrome Architecture and Sculpture in Antiquity (1834)
and Application of Colour in Architecture and Sculpture (1836) with the observation
that the arts, specific to their cultural and political context, formally transformed artistic
traditions while upholding elemental typal motives in idea (Fig. 5).

19
‘Giebeldecke vom Parthenon zu Athen’ [Pediment of Parthenon in Athens], in Gottfried Semper,
Anwendungen der Farben in der Architektur und Plastik (Dresden: Fürstenau & Co, 1836)

Semper’s search for the origins of architecture in typal motives was first articulated in
drafts for a never completed book, the Comparative Theory of Building (c. 1840s), and
later summarised in The Four Elements of Architecture (1851). They consolidated his
‘dressing theory’ (Bekleidungstheorie) and ‘theory of material transformation’
(Stoffwechseltheorie), through which he identified the four elements of architecture
(hearth, roof, enclosure, and substructure) and corresponding technical arts (ceramics,
carpentry, textiles, and masonry).36 Stimulated by the anthropologist Gustav Klemm,
Semper’s work was essentially a cultural theory of artistic invention, explaining the
creative process as a modification of elemental artistic motives through stylistic formal
changes. This provided an explanation of the relationship between a typal concept and
typological articulation through a problem of artistic design that is not found in either
Quatremère or Durand’s work. It also developed a concept of abstraction that derived

20
from the creative design process itself. Semper concluded his comparative theory of
style first in his London writings, in the early 1850s, and a series of lectures held at the
Department of Practical Art in London from 1853 to 1854.37 It was also in London that
Semper saw his thesis of the four elements of architecture represented in a ‘Caraib
Cottage’ from Trinidad, which was displayed at the Great Exhibition of 1851 (Fig. 6).

‘Karibische Hütte’ [Caraib Cottage], in Gottfried Semper, Der Stil in den technischen und tektonischen
Künsten, oder praktische Ästhetik: Ein Handbuch für Techniker, Künstler und Kunstfreunde (Munich:
Bruckmann, 1863)

In his lecture ‘Outline for a System of a Comparative Theory of Style’ (1853), Semper
proposed a compromise between a typal and typological reading of culture and its
history, while formulating a theory of building based on a comparative analysis of
elemental types. Stylistic variations, so Semper, were the practical means of necessary
utilitarian-material transformations in the becoming of form, disintegrating formal
traditions and revealing in their abstraction a symbolic and cultural context. Style in this

21
way became important to the definition of a material model and made art relevant to
society. Semper demanded in the lecture that architecture should follow the modern
sciences and adopt a systematic ordering of knowledge through classification.
Recalling his visits to the Jardin des Plantes in Paris as a student, Semper argued that
the methodologies of natural history could equally apply to studies of the arts:

They are like those of nature, connected together by some few fundamental
Ideas, which have their simplest expressions in types. But these normal forms
have given and give rise to an infinite number of varieties by development and
combination […] Will it not be important to trace out some of those types of the
artistical forms, and to follow them in their gradual progress from step to step
up to their highest development? A method, analogous to that which Baron
Cuvier followed applied to art, and especially to architecture would at least
contribute towards getting a clear insight over its whole province and perhaps
also it would form the base of a doctrine of Style, and of a Sort of topic or
Method, how to invent […].38

Like Quatremère, Semper saw elemental types as conceptual diagrams of invention.

And appropriating Cuvier, he believed that methodologically a comparison of material
formation could clarify invention and its relation to problems of styles. This
‘comparative theory’ was already discernible in Durand, however, Semper understood
it not just as a graphic analysis in the service of formalism, but a means to analyse the
motivations of artistic production and their development, insights that could be then
applied to design in both functional and conceptual terms. He was also critical of
Durand’s conception of utility as simply a human function, as he understood it as
constantly transformed by changing ornament and technical production, adapting to
practical and social human need. Accordingly he declared that the industrial arts gave
birth to architecture, and style obtained importance in a work of art by ‘observing the
limits, which are contained in and defined by the task and problem in question’.39

Style motivates typological transformations and reveals underlying typal ideas through
its changing means of abstraction. For example, Semper’s main architectural thesis of
dressing illustrates a process of changing material abstraction, when the spatial
covering of wall and ceiling evolve from temporary textile screens into permanent and

22
solid walls. Throughout its material transformations, the artistic motive of textile
decoration was maintained, not in resemblance but in idea. To Semper this intrinsic
relationship between typal idea and typological transformation is also manifest in the
etymological roots of the German words Wand (wall) and Gewand (dress), and further
supported by archaeological evidence of a developmental link between Egyptian,
Assyrian, and Greek polychrome styles in architecture.40 This reading of style as a
conceptual problem of abstraction and transformation that effects materiality,
essentially conforms to one given by Quatremère as ‘that which is least material, that
is the conception of ideas and the art of developing them according to a certain order’.41
Semper, however, focuses on the interrelationship between type and style. Types are
to him abstract and necessary forms, whose first materialisation is always modified
into new forms of artistic abstractions.

The Styles, which then resulted out of these secondary treatments were
composite Styles, which partook on one hand of the types, and the conditions
of Style, of the old materials employed for the latter, and on the other hand, they
partook of the Style which suits the new selected substance and manner of
treatment.42

Following a succession of material transformations over time, the effects of type and
style become hybridised, with the original artistic motive realised in different materials.
Accordingly, the same material has to be able of accommodating different stylistic
abstractions. Semper’s doctrine of style explicates how the translation of a generic idea
into a specific form can be conceived by conflating Quatremère’s metaphysical and
cultural idealism with Durand’s deterministic and utilitarian materialism. He develops a
synthesis between typal and typological abstraction in which formal invention is a
precondition and an outcome. Consolidating theory and practice, he provides an
example of how an abstract, theoretical type can provoke continuous formal
interpretations and material transformations, and how a typological comparison makes
form available to rigorous analysis. In this sense, his architectural theory conceives
form as arising from the combination of social and formal or material diagrams.

CONCLUSION

23
Quatremère, Durand, and Semper’s theories are consistent with the principles of
transformative composition in rhetoric. Quatremère’s theory of type articulates the first
and indispensable canon of ‘invention’ (inventio) by establishing a systematic
architectural theory of invention that defines the disciplinary means and principles
through which coherent arguments are generated in practice.43 Durand’s method of
design in turn is based on the second canon of ‘arrangement’ (dispositio or taxis),
which follows once an argument or idea is strategised by invention. Arrangement
manages the relative and iterative ordering of the part to the whole and organises
arguments into an effective discourse stating, outlining, and providing proof for a given
case or problem. Finally, Semper’s doctrine relates to the canon of ‘style’ (elocutio) by
discussing the appropriate and effective modes to express ideas. Whereas invention
determines what is articulated, style articulates how it is communicated.

The three theories considered problems of historicity in the architectural work and
proposed a resolution through abstraction: conceptually, diagrammatically, and
materially. Their mobilisation of history profoundly changed the conception of
architecture and revealed type as only conditionally autonomous, at the very moments
when through the translation of typal ideas disciplinary knowledge is challenged,
changed, and enriched. Prompted by an eighteenth-century transformation of practice
and systematisation of theory, architecture became a modern discipline with its own
claim to a specific knowledge, which only became possible by distinguishing it from
its history. The advent of the notion of type in architecture in the early nineteenth
century was instigated by a pervasive obsession with origins and fundamental
advances in archaeology, art history, anthropology, etymology, grammatology, and
zoology. This highlights that type and typology are unspecific to architecture and
interdisciplinary ideas through which knowledge is ordered and obtained. But the
instrumentalisation of type and typology in architecture discloses their conception in
didactic terms, how theory and practice are imagined as an indivisible material and
social construct.

The eventual demise of type in the twentieth-century architectural discourse

encouraged a turn to diagrams. However, as Le Roy recognised, architectural
diagrams rely on typological production. He also understood that a diagrammatic
function depends on abstraction, an abstraction of architectural form that considers

24
history, context, and culture as discursive arguments and limits. The typological
diagram therefore simultaneously envisions architecture as a specific object and a
generic possibility of objects. The diagram limits the possibilities of architectural
speculation without determining a finite formal representation. Although diagrams are
instrumental to arrange and convey relationships, as the theories of type reveal,
diagrams are only intermediaries between conceptual thinking and material
representation. Diagrams contribute productively to a tension between type and model
by offering a possible synthesis. Accordingly, a formal individuation is a clarification of
a typal idea through the translation of a diagram into a possible material manifestation.

By abstracting common organisational and structural diagrams of architectural cases,

type can be analysed and projected, as anticipated by Durand and clarified by Semper.
This presents, on the one hand, formal solutions receptive to transformation and, on
the other, represents a repository of knowledge. Thus, the utilisation of diagrams is
essential to conceptualise and analyse form. Although formative to the historical
discourse of architecture, types do not require continuity, as evident in Semper’s
theories, and are defined by transformations through which they effect change. As Alan
Colquhoun wrote, to ‘understand any given cultural situation, we must investigate its
synchronic structure rather than try to explain it exclusively in terms of diachronic
development. The synchronic situation always contains traces of the past’.44 Types are
an integral part of the physical-material, socio-political, symbolic-cultural, and historical
conception of our cities and their architecture, but a typal and typological reasoning
never just looks at the past and is directed towards the present. While the questions
arising with the concepts of type and historicity have defined a modern reasoning of
architecture, this was not to establish static norms, rather to advance continuing
practice and knowledge. As Semper insisted, artistic progress, despite evolving from
past traditions, becomes only possible when traditions are disintegrated by
contemporary culture.

1 Compare formative texts such as Stan Allen’s ‘Diagrams Matter’ (1998), Peter Eisenman’s ‘Diagram: An Original

Scene of Writing’ (1999), and Robert Somol’s ‘Dummy Text, or The Diagrammatic Basis of Contemporary
Architecture’ (1999).
2 See in particular Peter Eisenman, ‘Diagram: An Original Scene of Writing’, in Peter Eisenman, Diagram Diaries (New

York: Universe Publishing, 1999), p. 31.

3 A well known example of this normative reduction of building types is Ernst Neufert’s Bauentwurfslehre (Architect’s

Data), first published in 1936.

25
4 Jeffrey Kipnis, ‘Re-originating Diagrams’, in Peter Eisenman: Feints, ed. by Silvio Cassarà (Milan: Skira, 2006), p. 196.
5 Carlo Giulio Argan, ‘Sul concetto di tipologia architettonica’, in Karl Oettinger, and Mohammed Rassem, eds,
Festsschrift für Hans Sedlmayr (Munich: Beck, 1962), pp. 96–101. English translation: ‘On the Typology of Architecture’,
trans. by Joseph Rykwert in Architectural Design, 33 (1963), 564–65 (p. 565).
6 Argan’s article has to be seen in context to Ernesto Rogers’s ‘Preexisting Conditions and Issues of Contemporary

Building Practice’ (1955), which formulates the concepts of continuità, and Saverio Muratori’s morphological study of
Venice in Studi per una operante storia urbana di Venezia (1960). They influenced the extension of typology to morphology,
and a new rationality connecting architectural design strategies to that of the city. This is evident, for example, in
discussions by Carlo Aymonino and Aldo Rossi of typology and morphology in 1963–65, and subsequently in Rossi’s
Architecture of the City (1966), Vittorio Gregotti’s Il territorio dell'architettura (1966), and Giorgio Grassi’s La costruzione logica
dell'architettura (1967).
7 Werner Oechslin, ‘Premises for the Resumption of the Discussion of Typology’ in Assemblage, 1 (1989), 36–53 and

‘For a Resumption of the Typological Discussion’ in Casabella, 509–510 (1985), 66–75.

8 See Micha Bandini, ‘Typology as a Form of Convention’, AA Files, 6 (1984), 73–82 (p. 75).
9 These include Anthony Vidler, ‘From the Hut to the Temple: Quatremère de Quincy and the Idea of Type’, in The

Writings of the Walls (New York: Princeton Architectural Press, 1987); ‘The Idea of Type: The Transformation of the
Academic Ideal, 1750–1830’, Oppositions, 8 (1977); and ‘The Third Typology’, Oppositions, 7 (1976).
10 For example Argan’s ‘On the Typology of Architecture’ (1962), Rossi’s Architecture of the City (1966), Colquhoun’s

‘Typology and Design Method’ (1967), Vidler’s ‘The Third Typology’ (1976), Moneo’s ‘On Typology’ (1978), and
Oechslin’s ‘Premises for the Resumption of the Discussion of Typology’ (1989).
11 Compare the most recent, comprehensive reassessments of the work of Le Roy and Quatremère by Christopher

Drew Armstrong, Julien-David Leroy and the Making of Architectural History (2012) and Sylvia Lavin, Quatremère de Quincy
and the Invention of a Modern Language of Architecture (1992).
12 Claude Perrault, Ordonnance for the Five Kinds of Columns After the Method of the Ancients [Ordonnance des cinq espèces

de colonnes selon la méthode des anciens, 1683]. Translated from French by Indra Kagis McEwan (Los Angeles: Getty
Publications, 1993), p. 52.
13 This can be compared to Anthony Vidler’s description of history and theory in Johann Joachim Winckelmann’s

work: ‘The first was the body of rules governing taste, form, composition, propriety, and genre, generally understood
as the ‘theory of art’. The second was the account of the remains of the past commonly understood as ‘history’,
sometimes chronological, sometimes heterotypical, sometimes arranged by juxtaposition, similarity, or resemblance.’
In ‘The Aesthetics of History’, in The Writings of the Walls, p. 127.
14 This thesis anticipates Winckelmann’s dogma in the History of the Art of Antiquity (1764) that all art historical enquiries

ought to consider the context of a work of art through the culture, climate, and geography of the society that produced
it.
15 Julien-David Le Roy, The Ruins of the Most Beautiful Monuments of Greece [Les Ruines des plus beaux monuments de la

Grèce: considérées du côté de l'histoire et du côté de l'architecture, 2nd edn, 2 vols, 1770]. Translated from French by
David Britt (Los Angeles: Getty Research Institute, 2004), p. 210.
16 Earlier comparative plates by Jacques Tarade, Juste-Aurèle Meissonnier, and Gabriel-Pierre-Martin Dumont, which

are likely to be known to Le Roy, are surveyed by Robin Middleton in ‘Introduction’, The Ruins, pp. 90–97.
17 Le Roy, Histoire de la disposition et des formes différentes que les Chrétiens ont données à leurs temples depuis le règne de Constantin

le Grand jusqu'à nous (Paris: Desaint & Saillant, 1764); as cited by Christopher Drew Armstrong in Julien-David Leroy and
the Making of Architectural History (London: Routledge, 2012) p. 162.
18 As Le Roy explains, ‘the forms of the buildings largely depend on climate and [that] the principles of architecture

are not all so general that they do not sometimes give way before such influences’. The Ruins, p. 229.
19 Michel Foucault, The Order of Things: An Archaeology of Human Sciences [Les Mots et les choses: une archéologie des

sciences humaines, 1966]. Translated from French (New York: Pantheon Books, 1971; repr. 1994), p. 221.
20 The cultural anthropologist Melville Jean Herskovits distinguishes in the formation of culture by individuals of a

society between socialisation (a formal social integration) and enculturation (an unconscious conditioning through customs
that can change). Cultural change taking place as a result of exchange between different cultures is defined as diffusion
(when a cultural transmission is achieved) or acculturation (when a cultural transmission is in process). See Man and his
Works (New York: Alfred Knopf, 1948; repr. 1956).
21 The Encyclopaedia, or a Systematic Dictionary of the Sciences, Arts, and Crafts (1751–72) by Denis Diderot and Jean-Baptiste

le Rond d'Alembert eliminated Charles Batteux’s category of the beautifying arts that previously included architecture
and instead positioned it alongside the fine arts of music, painting, sculpture, and engraving.
22 Antoine-Chrysostome Quatremère de Quincy, An Essay on the Nature, the End, and the Means of Imitation in the Fine Arts

[Essai sur la nature, le but et les moyens de l’imitation dans les beaux-arts arts, 1823]. Translated from French by J. C.
Kent (London: Smith, Elder and Co., 1837), p. 172.
23 Quatremère, Essay on Imitation, p. 21.
24 Quatremère, Essay on Imitation, p. 258.
25 ‘The words idea and image being synonymous, some metaphysicians have proposed to determine their variation, by

applying the word idea to notions of intellectual objects, and the word image to those of corporeal objects.’ Quatremère,
An Essay on Imitation p. 212.

26
26 ‘Type’, in The True, the Fictive and the Real: The Historical Dictionary of Architecture of Quatremère De Quincy [originally in the

Encyclopédie méthodique: Architecture, vol 3, 1825]. Translated from French by Samir Younés (London: Andreas
Papadakis Publishers, 2000), p. 254.
27 ‘Type’, p. 255.
28 ‘Type’, p. 255.
29 Legrand offered in 1799 a historical and theoretical text based on his planned work entitled Histoire générale de

l’architecture to Durand. The essay was included with some first editions of the Collection and Parallel of Buildings and in
1809 published separately.
30 Jacques-Guillaume Legrand, ‘Essai sur l'histoire générale de l’architecture’, in Jean-Nicolas-Louis Durand, Recueil et

parallèle des édifices de tout genres, anciens et modernes, new edn. (Paris, 1809), as trans. by Anthony Vidler in ‘The Idea of
Type: The Transformation of the Academic Ideal, 1750–1830’, Oppositions, 8 (1977), 94–115 (pp. 106–07).
31 Jean-Nicolas-Louis Durand, Précis of the Lectures on Architecture with Graphic Portion of the Lecture on Architecture [Précis

des leçons d'architecture données à l'École Royale Polytechnique, 1802–5 and Partie graphique des cours d'architecture
faits à l'École royale polytechnique depuis sa réorganisation; précédée d'un sommaire des leçons relatives à ce nouveau
travail, 1821]. Translated from French by David Britt (Los Angeles: The Getty Research Institute, 2000), p. 77.
32 Durand, Précis, p. 126.
33 This contradiction is pointed out by Leandro Madrazo in ‘Durand and the Science of Architecture’, Journal of Architectural

Education, 48.1 (1994), 12–24.

34 Durand, Précis, p. 180.
35 Jean-Nicolas-Louis Durand, Partie graphique des cours d'architecture faits à l'École Royale Polytechnique depuis sa réorganisation;

précédée d'un sommaire des leçons relatives à ce nouveau travail (Paris: the author, 1821). English translation: Précis, pp. 185–6.
36 Semper’s idea of the four elements was indebted to Gustav Klemm’s anthropological work the General Cultural History

of Mankind (1843–52), which coined a modern concept of culture as formed by society and its morals, laws, and
customs, and described the houses of the South Pacific as centred around the hearth, elevated on earthen platforms,
enclosed by walls made of wicker, and covered by a sparred timber roof. As Klemm claimed, art did not originate from
fixed types but a primordial human need for representation that expressed itself as an artistic drive: ‘We find the
beginnings of art in the lowest stages of culture, where we also encounter the beginning of nations, because man has
the urge to manifest his experiences externally, and to adorn his environment with these representations.’ In Allgemeine
Cultur-Geschichte der Menschheit, I, p. 214, as cited by Mari Hvattum, in Gottfried Semper and the Problem of Historicism
(Cambridge: Cambridge University Press, 2004), pp. 43–44.
37 See The Four Elements of Architecture (1851), Science, Industry, and Art (1852), Theory of Formal Beauty (late 1850s).
38 Gottfried Semper, ‘Outline for a System of a Comparative Theory of Style’ (1853). MS 122, fols. 5-6, in the Semper

Archive, ETH. Reprinted in Gottfried Semper, Harry Francis Mallgrave, and Joseph Rykwert, ‘London Lecture of
November 11, 1853’, RES: Anthropology and Aesthetics, 6 (1983), 5–31 (pp. 8-9).
39 Semper, MS 122, fol. 15, repr. in ‘London Lecture’, p. 11.
40 For the etymological argument see Gottfried Semper, Style in the Technical and Tectonic Arts; or, Practical Aesthetics, trans.

by Harry Francis Mallgrave and Michael Robinson (Los Angeles: Getty Publications, 2004), p. 248.
41 ‘Style’, in The True, the Fictive and the Real, p. 238.
42 Semper, MS 122, fols. 30-31, repr. in ‘London Lecture’, p. 16.
43 Quatremère’s library included La Rhetorique d’Aristote (Paris, 1822) with notes and an index of parallels in Cicero and

Quintilian; see Fournel, Bibliothèque de M. Quatremère de Quincy de l’Académie des inscriptions et belles-lettres, secrétaire honoraire
de l’Academie des beaux-arts (Paris: 1850), p. 27.
44 Alan Colquhoun, ‘Introduction: Modern Architecture and Historicity’, in Essays in Architectural Criticism: Modern

Architecture and Historical Change (Cambridge, MA: The MIT Press, 1981), p. 14.

27