Last week, the architecture department at the University of Cambridge was saved from the indignity of designing its own memorial. It had been under threat of closure since last November, when the university lost patience with its 4-star rated research profile and the resulting loss of £350,000 a year in research grant. In an act of academic solidarity, the University's school of arts and humanities last week agreed to three years' subsidy for the world-renowned department.

Last week, the architecture department at the University of Cambridge was saved from the indignity of designing its own memorial. It had been under threat of closure since last November, when the university lost patience with its 4-star rated research profile and the resulting loss of £350,000 a year in research grant. In an act of academic solidarity, the University's school of arts and humanities last week agreed to three years' subsidy for the world-renowned department.

"It's extremely good news," said Professor Marcial Echenique, head of the department. "The support of the public has been marvellous." But Echenique is aware that the lifeline only stretches as far as the financial settlement following the 2008 Research Assessment Exercise (RAE). The department's ultimate future is in the hands of the RAE panel. As he said before Christmas: "In reality, if we don't get a 5 or a 5-star, we'll be in serious trouble."

After watching the axe raised over Cambridge, that sentiment must be shared by other architecture schools in the research-led Russell Group. Like Cambridge, their funding models tend to use Hefce research grant as cross-subsidy for under-funded teaching budgets. But a vocational and practical subject like architecture, where research output is hard to assess and studio-based teaching cuts into lecturers' research time, risks becoming the cuckoo in nests feathered with research funding.

"Architecture does sit uncomfortably in research-intensive universities," agrees Professor Malcolm Grant, provost of University College London. "There is an inflexible discipline imposed by research funding when you're operating at the margins, as so many are." Grant speaks from experience: UCL is parent to the Bartlett School of Architecture, which suffered the same fate as Cambridge when it dropped from a 5 to a 4 in the 2001 RAE. As a result, research funding was cut by around £250,000 a year.

The Bartlett avoided the same fate as Cambridge because its flexibility to recruit extra students allowed it to make up the funding shortfall. But a UCL statement on the future of the Bartlett says that "We are seeking to maintain and improve academic quality across the board... and we will not want to sustain poorly-ranked activity." In other words, if UCL was ever forced to choose between its research reputation and its architecture school, it would reach the same conclusion as Cambridge did in November.

The architectural profession, represented by the Royal Institute of British Architects, is alarmed by the subject's vulnerability. "Unless we change the model of funding or improve the rankings of schools of architecture, we could find the subject is no longer represented in our top universities," says architect Jack Pringle, RIBA president-elect. "Of course we're happy that the newer universities are doing well in architecture. But we also want students who are attracted to the Russell Group, and who might study something else if architecture isn't an option."

The stress on research could unbalance architectural schools, which traditionally have one foot in academia and one in architectural practice. "The effect is that it's harder to employ someone who works in an office and doesn't have a PhD," says Professor David Dunster, head of the school of architecture at Liverpool University, another Russell Group member. "You're getting a gap between what people are researching and what's being taught in the studio [by part-time practising architects]. Splitting off teaching and research isn't really higher education."

Cambridge's response to its post-RAE funding crisis bears out Dunster's point. "We offered early retirement to people who are less research-assessable, who were mainly practising architects. We're taking on one new chair and one lecturer, and those people will be research assessable," says Professor Echenique. "The typical integration between design and research in architecture schools could be lost. We will need to make a tremendous effort to re-integrate them in our mode of teaching."

At the heart of the problem is the way the RAE treats architecture. Hefce doesn't straitjacket architecture into refereed papers or technical research: practical work, competition entries and exhibitions can all be put forward. But no credit is given for a design concept or project which evolves in a university studio but takes on a practical application some time later.

The RIBA's Pringle would like to see research encompass "everything that moves the profession forward, including exploratory designs that develop an energy which works through into practice." Kate Heron, head of the architecture school at the non-Russell University of Westminster, agrees that "It's a pity that practice doesn't count as research - everyone would benefit from that."

There were also issues with the methodology and personnel of the 2001 RAE. Architecture was treated as a sub-set of the Built Environment rather than a subject in its own right. The 13-strong panel was chaired by a surveyor, dominated by technical and construction experts, and included no practising architects. The inference is that the panel over-weighted technical research at the expense of design work.

Lobbying from the RIBA and schools appears to have had the desired effect. In 2008, the relevant HEFCE sub-panel will examine "Architecture and the Built Environment". The profession is confident that the panel's chairman will be an architect. A promisingly wide list of organisations in the field will be consulted over the panel's membership. "With better representation, the full spectrum of criteria will be brought to bear," says Pringle.

The 2008 RAE will be crucial, as the panel members themselves will be aware. "If the panel gives a relatively low grading to a school in a research-intensive university, there's every chance they're signing a death warrant," warns UCL's Provost Malcolm Grant. But he adds: "It's not inevitable. The universities and the profession have to press the case of research and innovation in architecture as a serious object of inquiry."

At Sheffield University, also in the Russell Group, head of architecture Professor Jeremy Till makes a similar point. "Cambridge gives out the signal that architecture isn't a serious academic discipline, so architecture needs to be clear about its intellectual credibility. [Cambridge] isn't a death-knell, but it is a wake-up call that we need to see architecture as a serious research discipline."

education@independent.co.uk