Theory runs into a building

When I was a student of architecture, I argued with my tutors about many things (this will surprise no one), but especially about the teaching of construction. Although, even then, I loved architecture for its ideas, my struggle - and this is what may surprise you - was to be taught building; bricklaying, carpentry and welding.

I didn't want just to know about these things. By second year I could draw the details, pass the exams, win the prizes. I wanted to know it with my fingertips. I craved what Renzo Piano, I would later discover, calls the ''language of your stomach''.

No, stonewalled my tutors. To a man, practitioners as well as theorists, they invoked the need-to-know defence. ''Nuh. Don't need it. We design; builders build.''

I thought that was fundamentally wrong. Still do. Not because I was interested in building. Heavens, no. The orthodoxy of joists, perpends and damp-proof courses held no enchantment for me. Not even because I believed that architecture is in essence a craft (rather than an art or scholarly discipline), although there is some truth to this view.

No, I harangued my teachers for real-life on-site construction lessons precisely because I loved the architecture's head-stuff. I was in love with its ideas and felt certain - from a point of almost complete ignorance - that the best of them were rooted in real, material stuff.

The placing of one brick on top of another may be deeply, inherently, dull, but (I felt) it is only from that knowledge that the poetry springs.

This belief, unlike most student foibles, persists. In the theory stakes it inclines me toward the Moderns like Corbusier and Aalto and proto-moderns (such as Semper, Ruskin, Viollet-le-Duc and Gaudi) who believed matter should inform the intellect, as well as vice versa. It's not theory, but it points the gap to the theory-shaped hole at the heart of architectural practice.

Much has been said of late about the death of architectural theory, most of it by architectural theorists. After the treatises of classical antiquity and the manifestos of modernism, we live, it is said, in a ''post-critical'' age. Thousands of theoretical books and papers are still written, and lectures grandly delivered. Yet our built world marches on, unchanged, oblivious.

So what is architectural theory? Does it, in fact, exist? In many disciplines the role of theory is clear. In physics, it speculates beyond the bounds of knowledge in order to extend those limits ever outward, widening the circle of the firelight.

Theory is more problematic in the arts, although in music it seems reasonably well-bonded to practice; from the ultra-esoteric, on one hand, to third grade kid stuff on the other. Imagine if we taught our children architecture theory in that way, in schools, hand-in-hand with practice and composition. What would the lessons contain?

The first architectural theorist is usually taken to be Vitruvius, who lived and died BC. His Ten Books provide little enough succour for the contemporary designer but they do traverse a considerable spectrum, from symbolic obligation (which way to face your temple) to practical know-how (how to find and channel water). For 1800 years, theory did little more than translate, elaborate, interpolate and illustrate Vitruvius. Then the classical straitjacket became intolerable and Modernism burst in.

Modernism was a new world. Chanelling Marx, it no longer produced treatises but manifestos; from the 1909 Futurist Manifesto, the 1918 de Stijl Manifesto to the 1942 CIAM Athens Charter. The content varied, except for one conviction: that architecture wasn't just an elegant carapace for human life. It was a blazing moral torch.

Modernism's torch burnt for half a century, off and on, until smothered beneath the thick wet blanket of French theory. Foucault, Barthes, Derrida, Lacan and Gilles ''the fold'' Deleuze persuaded architecture to self-flagellate with knotted whips of jagged long-chain jargon. And thus we remain, hurting, gasping for air.

Professor Anthony Vidler, the dean of architecture at the revered Cooper Union in New York, argues persuasively that architectural theory has, by these disciplines, been breached, colonised and destroyed.

Theory, he says, was once ''a means of thinking about architecture from inside the discipline'' but has become so dominated by external voices - from philosophy, linguistics, psychoanalysis and sociology - that it is now entirely hollow-core.

Which is my point, really. French theory was always long on critique but short on creativity. The critique, however, diminished by the profound and often willful obscurity of the writing, has nevertheless been genuine and useful, sensitising architecture to context, perception, sustainability and social responsibility. Sooner or later, though, the moment of architecture arrives; the moment when the line must be drawn.

Where that line goes, and how it relates (or not) to other lines, is the essential architectural act. And it is here, where other disciplines and exigencies end, that theory should come into play.

Functionality may demand a room here, or a window there. It may even require certain qualities of that room or window. But in determining what kind of room or window, theory should be available to guide the hand according to the building's intellectual core.

For classical or modern architects, the appropriate theoretical protocol - be it the classical orders, the golden mean, or Corbusier's ''modulor'' - was always at hand, ensuring coherence throughout, from floorplan to door handle.

But where theory is absent, other exigencies - digital parametrics or dumb-ass bean-counting - dominate, and we end up with a world that looks like it was designed by bored, cyber-addicted project managers.

If we are to control the emerging geometries of the digital era, argues Vidler, we must rebuild ethical and formal judgment. Theory must emerge, blinking, from the echo-chamber where it speaks in riddles like some narcissistic Sibyl. It must re-address the world.

Theory must become intelligible and purposeful, connecting materiality with mind, linking head, heart and hand. It must, in short, guide the line onto the page. Me, though, I lost my construction argument, so I still don't know my Flemish bond from my Raking Rat-trap. Lucky I was always handy with a comma.

Dr Elizabeth Farrelly will chair a free public debate, ''No Such Thing as Architectural Theory'', at 4.30pm as part of the UNSW LuminoCITY exhibition at Pier2/3 Walsh Bay.