It might seem odd for the government to lavish time and money on a building that is, after all, primarily a pit stop for dead bodies. But the mortuary, which opened in September 2005, also houses some of the city's most formidable forensic resources - and for a relatively safe and healthy place, Hong Kong has no shortage of mysterious, sometimes gruesome fatalities requiring in-depth investigation. In the past few months alone, a dismembered torso surfaced off the Central Star Ferry pier, two visiting American businessmen perished of apparent drug overdoses in a five-star hotel room and a young Indonesian worker was stabbed and left to die at dog kennels in Fanling.

When unsettling cases like these arise, the public eye is inevitably fixed on the victims, police or suspected killers, but it is often the city's forensic pathologists who come up with the evidence needed to convict the guilty. The Department of Health has a team of 16 experts, who kick into action every time someone passes away under questionable circumstances or in official custody. They pore over places of death, inspecting bodies for telltale injuries, and report their findings to the police or courts. It is a grim but essential calling that, in the words of Hau, allows the dead to speak - to help the living.

For an idea of the impact of forensic work, one only has to delve into the casebook of Dr Philip Beh Swan-lip, a former government pathologist who is now an associate professor at the University of Hong Kong's Department of Pathology and one of the city's foremost experts in the field. By his own estimate, Beh has conducted more than 12,000 autopsies, a count that seems to have taken little toll on the smiling, affable academic. He says his occupation has given him 'a very philosophical view of death'.

'I've seen enough cases to know that there's no way you can predict when it's going to come,' he says, laughing.

But Beh has also seen plenty of instances in which deaths were deliberate - and has made sure some of the people responsible have been held to account. In his long career, a few such cases stand out, such as one that began two decades ago when the police found a female torso - only a torso - in a nondescript flat.

'It was memorable because Hong Kong doesn't have many cases involving dismembered bodies,' Beh says. And, no doubt, for the work that followed.

The exhaustive search for the victim's missing limbs led the police, and the professor, to the massive dumpsters serving the housing estate on which the victim lived. They tipped over the dumpsters and sifted through the contents in the dead of night. The team eventually found the limbs, but more importantly, multiple bone fragments that forensic experts painstakingly reassembled to form a skull. 'It was totally crushed,' recalls Beh. 'Some of her teeth were found, but ... we couldn't actually complete the face. In those days the whole notion of facial reconstruction was still very new.'

So was DNA testing, which meant Beh, the police and dental experts had to work for weeks to formally identify the body, which turned out to be that of a missing schoolgirl. A significant lead on her killer came a few months later, when police found a glass with a trace of her fingerprint on it in a neighbour's apartment.

'From what we gathered in the end, he was trying to fill in some forms but didn't know how to and somehow managed to get this girl to agree to help him,' says Beh. 'She ended up with him in the flat and he offered her a drink. Something happened, she screamed and he blew a fuse because he associated the screaming with his wife and daughter, who had just abandoned him.' In his desperation to dispose of the body the murderer cut it into pieces and even attempted to cook some parts.

'It was one of those strange cases where we don't think there was any premeditation; she was unfortunately in the wrong place at the wrong time. And he was eventually convicted of manslaughter, because we couldn't really prove he planned it,' says Beh.

Hong Kong's forensic professionals are sometimes called to weigh in on high-profile investigations, such as the one into the death of Annie Pang Chor-ying, a one-time model whose headless skeleton was discovered in a Yau Ma Tei apartment in 1999. Pang had been missing for four years.

Suspicions initially swirled around the apartment's owner, John Fang Meng-sang, a lawyer and brother of Legislative Council member Anson Chan Fang On-sang who was once Pang's lover. Fang hotly denied allegations he had ordered her death, but it was Hau's input at a 2006 inquest that saw a jury exonerate him. The pathologist's examination of Pang's remains was unable to produce a specific cause of death - Hau told the coroner's office that while suffocation or trauma were possibilities, there were no signs pointing to foul play. The fact Pang's skull was found in a rubbish bin next to the body could be explained by decomposition wearing down the tissue that bound it to the rest of her skeleton, assuming her neck rested on the lip of the bin. The pathologist also noted traces of opiates in Pang's hair samples, pointing to a possible drug overdose. All in all, his testimony was enough to convince the five jurors to classify Pang's death as an 'accident or misadventure', probably rooted in a history of narcotics use.

The most disheartening cases for forensic practitioners are those in which they can't provide any answers. Throughout the 1990s, overcrowded internment camps set up in Hong Kong to handle the mass exodus of 'boat people' from Vietnam periodically erupted into violence, keeping police and government pathologists busy. The aftermath of one such riot that Beh was called upon to investigate still haunts him.

It began as a tussle between a few North and South Vietnamese residents, but before long 'the whole camp was on fire', recalls Beh. 'There were multiple deaths, mainly women and children, because the adults basically decided the whole place was burning up' and rushed outside, barring the exit for the weakest among their number. 'It affected me because it was probably the first time I had to deal with such a large incident. I was constantly worried I had left things out. We spent a whole day just documenting where the bodies were then moving them to the mortuary. Because they were all burned beyond recognition there wasn't a lot we could do. Their belongings were scattered all over the place so we couldn't rely on those.'

In the end, investigators counted 24 corpses. Because refugees entering Hong Kong were not subject to medical examinations - which some campaigners insisted would constitute an infringement of their human rights - investigators had no data on the dead to compare any findings to. 'So despite doing all the autopsies we had nothing,' sighs Beh. 'We didn't know if the claimed children of a person were really their children, and there was no way to say who was missing after the event ... it was a failing.'

But despite the occasional setback, Beh and Hau agree on two things - forensic science has seen some incredible advances in recent years and Hong Kong has become a much safer city.

'Even 20 years ago, we had no sophisticated DNA techniques,' says Hau. 'When we started DNA profiling, that helped a lot in solving very difficult cases, either by identifying the deceased or identifying the culprit. The other big change is demystification. In the past, people would look at forensic medicine as something only doctors understood but now, with the expansion of information, especially over the internet, people have started to know and expect more.'

Hau has also experienced what the pathology community calls 'the CSI effect' - a growing number of armchair experts weaned on the US crime drama who are overconfident in the power of forensics. 'I'm not saying the show is telling viewers untrue things, but most of the time events are dramatised, so sometimes you'll encounter people with unrealistic expectations,' he says with a laugh.

According to Beh, in some ways the pathologists of the 1980s and 90s had a lot more work to do. Hong Kong's murder rate was higher and investigators regularly had to examine the victims of botched armed robberies and triad turf wars.

Nowadays, in contrast, murder is largely a domestic affair, with more than 90 per cent of murders taking place in the home. Rather than machete-wielding gangsters, killers are likely to be spurned lovers, jealous spouses or parents who take the lives of their children out of depression and hopelessness.

The city's murder rate may be lower but Beh doesn't see that as cause for celebration. Over the past two decades the number of domestic murders has remained virtually constant. 'Imagine this; if we could prevent domestic homicides, we may be down to 10 a year or even less,' says Beh. 'That would really make Hong Kong a safe city.'

Beh believes one of the first steps towards addressing domestic violence in Hong Kong is better understanding, and he is assembling a database of local murder cases that classifies them according to criteria such as the cause and circumstances of a death. Although Beh's project has run into some resistance from tight-lipped government departments, it's already helped him catalogue some interesting trends. The bulk of murders here take place on weekends or early on a Monday morning - understandable, Beh says, because that's when families are likely to be at home together. 'When they're working, there's not enough time to argue,' he says. 'But over the weekend pressures can build to boiling point.'

Also, the average murder-suicide in Hong Kong exhibits traits different from those in Europe and North America. The culprits and victims are younger, their incomes usually lower and there are more cases here of 'multiple kills', in which people take the lives of their partners and children before their own. To Beh this suggests a clear course of action.

'When we look at what we need to do to alleviate these cases, we have to consider how much we know about our neighbours and whether social welfare is really targeting the right groups of people,' he explains. 'They may be focusing on the welfare and development of a child, ignoring the fact dad or mum, who has been faithfully looking after the kid for the past six or seven years with no respite, may be on the verge of a breakdown. The holistic nature of that scenario is sometimes missed.'

Unfortunately, if current trends continue, Beh will have less data from which to derive his assertions. Most forensic work is conducted on behalf of the government coroner, who receives reports on about 25 per cent of the city's deaths and decides which warrant further investigation. But in recent years, fatalities have been attracting less official attention. The ratio of referred cases for which the coroner has ordered autopsies has steadily declined, from 55.8 per cent in 2001 to 38.1 per cent last year. The number of deaths receiving police investigations has also dropped, from 30.7 per cent in 2001 to just under 12 per cent in 2006.

The most recent Coroner's Report attributes the shift to a couple of factors - first, the government has decided that when 'direct and co-operating evidence' clearly indicates a death was a suicide or accidental, the cost of a probe 'cannot be justified'. The coroner is also receiving more requests from upset relatives of the deceased to waive planned autopsies, and given that many view such procedures with 'fear and abhorrence', he expects the trend to continue.

Despite the coroner's limited budget, a spokesman for the Hong Kong Judiciary, which oversees the Coroner's Court, insists enough resources will always be made available for it to 'discharge its duties fairly and effectively'.

These assertions do little to ease the worries of experts such as Beh. The professor notes that in decades past people weren't comfortable with autopsies either, but he feels the coroner was less likely to respond to pleas to forgo them. Now, he says, there's 'an undue and overbearing tendency' on the part of the government to avoid offending the bereaved.

The professor is not so troubled about murders being passed off as suicides or accidents - though he admits that's a possibility. More frightening is the prospect that the real causes of many suicides in Hong Kong are going unnoticed. 'There's just under 1,000 suicides per year, but if you look closely very few receive investigation orders,' says Beh. 'The concerns are those of suicide research. Were some of these people under the influence of drugs? If they were women, were there unwanted pregnancies involved? We simply don't know.'

Social workers also believe the government has become less likely to order investigations or autopsies even when relatives want them. Although the judiciary can't provide statistics on the rate of refused autopsy requests, Tim Pang Hung-cheong, spokesman for the Hong Kong Patients' Rights Association, says there's been a increase in the rejection rate recently.

'In some cases there may be doubts regarding the death, but the coroner is likely to be satisfied with an apparent cause of death, especially if it comes from a doctor ... perhaps due to a limitation of resources,' he says.

Pang is currently assisting three families who believe negligence by doctors or hospitals played a role in the deaths of loved ones but, because of inaction on the part of police or the coroner, remain uncertain.

In one case, a 44-year-old woman was diagnosed with cancer of the uterus and was recovering in the hands of a private specialist before suddenly experiencing breathing difficulties that led to her death. Doctors believe blockage of the pulmonary arteries was the cause. The Coroner's Court has repeatedly denied her husband's request for a comprehensive investigation into the death and he's now considering legal action against the doctors involved.

Another case involves a woman whose five-year-old son suffered a pulmonary haemorrhage during an operation at a public hospital. According to Pang, the coroner has agreed to look into the death after the police initially turned down the mother's demands for an inquest, but has yet to schedule the probe.

Hau insists the Department of Health is committed to providing 'quality service' to the people of Hong Kong, and that any death warranting the coroner's attention will be investigated. 'But to what extent it will be investigated is entirely at the coroner's discretion, based on the information available,' he adds. 'From an academic perspective, of course, it's desirable to have each and every case investigated as thoroughly as possible, but in reality we have to strike a balance.'

The pathologist believes forensics in Hong Kong have been held back by the state of the city's mortuaries, which, with the exception of that in Kwai Chung, are 'old and worn out'. Investigators don't only need more space to store bodies, but more sophisticated hardware and close ties with other agencies.

'We have to keep ourselves updated, not just with advances in scientific knowledge [but also] the modus operandi of culprits, and that's why we need to have a close working relationship with departments like the police,' says Hau.

The department has been pushing for more facilities like the ultra-modern mortuary in Kwai Chung but has run into 'strong resistance' from communities who are reluctant to have a charnel house on the doorstep.

Perhaps they can eventually be won over by the arguments of forensic pathologists who have spent a lifetime dealing with death. For Beh, death is 'just another face of life' while Hau sees forensic investigation as one of the only ways in which science can contribute to justice.