On 19 February 2001, a bag was found in London's Regent's Canal, containing the dismembered corpse of a 31-year-old woman. Paula Fields was a mother of three who had become involved in drugs and prostitution. Police eventually linked the crime to John Sweeney, who was captured in March 2001, after seven years on the run for trying to kill a former girlfriend, Delia Balmer. Thanks to new DNA evidence, Sweeney was linked to another crime: the 1990 murder of an American woman named Melissa Halstead, whose body was found in the Westersingel canal in Rotterdam. On 4 April 2011, Sweeney was given a "whole life" sentence for the murders of Fields and Halstead.
'Blood, saliva, little bits and pieces ? I?m not going to miss it.' Photograph: Andy Hall
The anticipation kicks in as soon as I get the call. I get in the car, put on some classical music and start thinking. Every crime scene is different. It's nowhere near what you expect.
When I arrive, I get a briefing from whoever's at the scene. You listen but you don't necessarily agree. I call it ABC. Assume nothing. Believe nobody. Check everything.
Maybe it's the first time a police officer has seen a body. They'll give a fantastic description of a decomposing corpse, but I can see that with my own eyes. I'll ask: did you smell anything? Was there a window open? How did you kick the door in? Was there a key? Did the key come flying out? I'm trying to get them to see further than the body on the floor.
When I arrive at a scene, it's my thinking time. It's now sterile. The circus hasn't arrived, so I get suited up and go in with my notepad. What am I seeing? What am I hearing? I have a quirk: I tend to follow the left-hand wall around a room. If you go to Hampton Court maze and follow the left-hand wall, you get to the middle. It's a good technique. Blood distribution, saliva, little bits and pieces ? I'm not going to miss it. If the kettle's warm, if the window's open. Has the central heating kicked in? Which lights are on? Which are off? Has the toilet been flushed? Is the seat up or down? You may not know the relevance, but take in the details ? a ring of dust, an open drawer.
I can remember virtually every job I've done ? about 400 murder scenes and 200 suspicious deaths. When it comes to cordons, I always think big. Think meningitis and hope it's a cold. A cordon can always shrink afterwards. But people have to get on with their lives, so you can't cordon off a whole estate. How am I going to manage the situation without compromising the evidence?
With the Paula Fields case, you've got five bags in a canal, two entry points. Questions arise: who's going to carry five bags? How many people? It turns out it was one person. How heavy are the bags? Has he brought a car here? In this case, the bags were weighed down with brick and tiles.
I noticed a skip nearby. Perhaps that should be part of my scene. Most importantly, I asked myself, how am I going to deal with these body parts? I've got people from nearby offices gawping out the window at what I'm doing. All these things run through your head. It's all moving in real time.
You have to make a decision as to what you're going to get out of a scene, how much of the detail you're going to look at when you're at the mortuary or the laboratory. You need to be able to effectively reconstruct the crime scene at the court afterwards.
Maybe 70% of what you retrieve is not relevant. That doesn't stop you from finishing with a fingertip search, looking for that last piece of detail. You retrieve and work out the relevance later.
I'm the conductor of an orchestra. I don't need to know how to play the violin, but I need to know what sound they make and, more importantly, when to bring them in; to recognise when I need a blood pattern analysis expert, a forensic archaeologist, a pathologist, biologist or palynologist, which are pollen experts, an environmental scientist, a person to construct immersive, 360-degree images of the scene ? say, if you want to see where the petrol was splashed at a fire scene. I have to bring them to the scene at the right time and usually deal with egos. They're experts: they come to a scene and think they've got answers to everything. They haven't. They're just a cog in the machine.
On the canal bank, I asked myself: what am I going to do with these bags? Can I get a pathologist to the scene now? When is the post-mortem going to happen? If you take a limb out of an anaerobic situation ? where there's no air, so it won't decompose ? you can see the flesh going blue in front of your eyes. These things all add up in your mind. We should do something here, now, I decided. We got a big-scene tent in and got started.
We look for finger marks, hairs, fibres, obviously, but in this case we also looked at the plastic bin liners that held the remains. If you find a roll of plastic bags at a suspect's address, you can prove where the bag has come from.
You need to know this sort of thing. You've only got one chance. If you seize the wrong things to start with ? no matter how well you manage the people, the budget, the egos ? then you're lost.
'You don't want to think about something going wrong.' Photograph: Andy Hall
You're underwater, and inside your mask the sweat is dripping into your eyes, but you're thinking: got to keep on. No matter how cold or uncomfortable you are, you must get the problem solved. We have certain skills, certain equipment, but it still takes a special sort of person to look for a body in nil visibility.
We normally work for about an hour and then concentration begins to wane. If it's really cold, you do a shorter dive. When you're underwater, you're working hard drawing in air. You don't want to think about something going wrong. So you have faith in the team and in your equipment, even when you're gasping. It can become quite claustrophobic.
In water, the biggest danger is drowning. Then there's the barbed wire, the razor wire, the sharp knives, the obstructions. There's an area in the London docks that's got shopping trolleys piled on top of each other. If you're at the bottom, you've got eight or nine metres of shopping trolleys teetering above you. You're aware of the metal, aware of the danger.
If you're the guy who finds the body, then you've succeeded. Sometimes, one diver has come up and the next diver's gone down and found it on the first sweep. We all want to be the diver who finds it. We call them "Gucci dives".
Divers aren't just used for work in water; we're brought into anywhere breathing apparatus might be necessary ? drains, sewers, even the tube train after the 7/7 bombs. It was a small gap and very hot work. They were still looking for body parts.
In another situation, a knife had been thrown in at high water. At low water, we were confronted with a lot of mud, probably 2-3ft deep. We were all in line, searching for this knife with our fingertips. It was the worst sort of mud ? exhausting to move through because you can't get a proper grip. We found the knife, but I had a week off afterwards. I was exhausted, but I'd taken in a bit of sewage as well.
You'd be surprised what we talk about at times like that ? recipes, the best way to look after chickens. Some of the lads are really into self-sufficiency, so we have great conversations about edible mushrooms. We don't talk about football very often. It's a small team, but we're close, almost like a family.
In the Paula Fields case, two boys went fishing and when they fished out a bag, they didn't understand what they had. A member of the public walked past and understood
. The boys thought they'd found a piece of pork. What they had actually found was the pubic area of a female.
That was a Monday night. It was getting dark by that stage, so we returned the next day and for some reason the canal had gone absolutely clear. Usually it's silty and muddy-coloured, but on that day you could see the bottom. There were five other bags in the canal that the crime scene manager was interested in.
The canal is about 8ft deep. We used what we call a floating jackstay ? a rope tethered at each end that we lay out to guide us. It floats on the surface, so the diver can go all the way along it, holding it in one hand while looking. Then we did a fingertip search along the bottom. Your fingertips become so important. You touch something and you know it's worth more investigation. It's almost like a sixth sense.
We were looking for more bags. We didn't know if there were more parts missing. There are a lot of heads around in London just waiting to be found.
If you are searching for human remains, you've got to be respectful, especially if you're in a public area. We have a body tray with a cover on it, so, once we've located the body or body part, we'll strap it on out of sight. In a best-case scenario, the most anyone will see would be an orange stretcher with a white cover. The worst case would be a body bag.
In the case of Paula Fields, the bags were taken into the forensic tent. We didn't look in them. We just handed them over.
'There's a lot of nightmare potential.' Photograph: Andy Hall
Paula Fields was murdered in 2001; Melissa Halstead's remains didn't come to us until 2009, even though she was killed earlier. I got involved to see if I could link what happened to Halstead to the marks from Paula Fields.
We wanted to see if there were links: was the same type of saw used? Had the bones been sawn up in the same way? Sometimes a person will cut through the radius and ulna together. Sometimes it's clear they've cut through one and then cut through the other.
We receive bones from the pathologist. They usually remove most of the flesh. One end is cut by the pathologist, one end by the offender. How we proceed depends on fitting in with the wishes of the family.
To some extent you've got to detach yourself from the fact that this is a piece of human body. You've always got that in mind, and the sensitivities behind it. But if you really started thinking about it, after 30-odd years in the job, you'd realise you've looked at lots of pieces and there's a lot of nightmare potential.
In this job you need to have good visual awareness, pattern recognition. It's the ability to put together what's been going on.
When a person saws through bone, the blade will go through at different angles. Usually, someone is sawing through flesh as well as bone. It's moving. It's going to grab at the saw and pull it at all sorts of angles. And, of course, how is a person holding it down? It's not like they have it secured in a vice. I suppose some people might hold it in a Workmate, but most of the time it's on a table or on the floor or in the bath. They're not going to be cutting through smoothly and evenly.
Usually you get a whole series of lines going across the ends of the cut, from all sorts of angles. One of the first examinations is to look for paint or other material in the cut. We've got a lot of hacksaw blades here to consult. Blades of the same type, or from the same firm, will have a colour. Often we are able to confirm the paint on a saw or eliminate another. In this case, we had a yellow-painted hacksaw blade, but when my predecessor examined Paula Fields, there was blue paint on the ends of some of the cuts.
We're using microscopes that go to 40 times magnification. It's not just paint flakes; the paint actually smears on to the ends of the bone. You'll also see partial saw marks where they've given up.
In this case, there were two different saws used on both bodies: a hacksaw and something more like a panel or tenant saw. A heavier-duty saw leaves a fairly wide mark, a hacksaw leaves a narrow mark, so we look at tooth spacing. You may see little indentations at the bottom of the cut. In this case, on one of them there were 24 teeth per inch ? a hacksaw; and on the other were 10 teeth per inch ? a wood saw.
When the edge of a tool moves across the surface of bone, you get microscopic imperfections, even from a new tool. All tools will be slightly different. As it gets worn and damaged, it will leave more and more detail: more hints, more clues.
'The head, hands and feet have never been found.' Photograph: Andy Hall
I joined the investigation team in 2005. Not long after, the phone rang. It was a Dutch officer from Rotterdam. He said, "I need to speak to someone because we've had a DNA hit on a body and it concerns one of your cases."
Paula Fields had come down from Liverpool in about 1999 or 2000, and met Sweeney around September or October 2000. He lived at an address in north London, and she happened to be walking the streets in that area. It's accepted they had an affinity because they were both from Liverpool. She began a relationship with Sweeney and, with her drug habit and her lifestyle of choice, it attracted a bit of attention to where Sweeney was staying. Clearly he wasn't comfortable with that, because he was on the run from the police for a near murderous assault on a previous girlfriend, Delia Balmer, in 1994. Sweeney enjoyed Paula's company ? this is only my interpretation ? but it all got a bit too much for him. She'd steal from him, presumably to buy drugs. The dealers were coming round and Sweeney was having to fork out money, which would have annoyed him greatly. This is primarily what we consider to be the motive for the murder of Paula.
In December 2000 ? the exact date was never established ? she disappeared. In February 2001, some lads fishing in Regent's Canal noticed a bag. It contained what we now know to be body parts of a female. Divers found five further bags containing body parts, which we know belonged to the same female, minus the head, hands and feet, which have never been found. DNA analysis was carried out which revealed that the body parts belonged to Paula. Inquiries led them eventually to Sweeney's address.
John Sweeney was arrested in March, up in the City, on a building site, a stone's throw from the Old Bailey. He was found in possession of firearms, one of which was in his tool chest at work. He had a couple of shotguns at home.
At that time, the inquiry team went back to see Delia. She'd told police in her statement that Sweeney had confessed to her, earlier in 1994, to killing his girlfriend Melissa in Holland. Delia's information said that he'd killed her, chopped her up and thrown her in the canal network in Amsterdam. The Amsterdam police did in fact have a missing female, found in a canal in 1992. The investigators had a suspicion it was Melissa, but when they obtained DNA samples from Melissa's family. these did not match with DNA from the body parts. That body has never been identified, and this is where the investigation stalled.
Sweeney was arrested for the murder of Paula, and arrested and charged for the firearms offences, and also for the attack on Delia. But the Crown Prosecution Service decided there was insufficient evidence to charge him for Paula's murder. He was given a life term, but that meant he would have been released in around April of this year.
But in 2005, a Dutch cold case team had picked up an unsolved case from 1990, when a dismembered female body minus head and hands had been recovered from a canal in Rotterdam. They rang the historic forensic samples lab and said, "Do you have any samples
;from the postmortem of this unidentified female?" As luck would have it, almost at the back of a drawer, they found what I believe to be a vial of blood. The lab constructed a full profile, loaded it on to a missing persons database and discovered a full DNA hit with Melissa's family.
They exhumed the remains of the 1990 body in early 2008, took the relevant DNA samples and it was confirmed to be Melissa Halstead. Sweeney had confessed to dumping her body in Amsterdam, not Rotterdam, but that could easily have been Delia in her stress not remembering. Or else Sweeney was playing a game. Sweeney was in prison. There wasn't an intense pressure of time. There wasn't a serial killer on the loose, but there was a possibility he could come out of prison. He was the prime suspect. We just had to make a solid forensic case.
'So much can be seen? It's like magic.' Photograph: Andy Hall
There's a lot of printed stuff nowadays from computers, which is difficult to track down, but people do still write things down. In this job, there are handwriting comparisons, signature comparisons, questionable cheques, altered documents. Maybe ballpoint ink has been scrubbed or obliterated. Maybe the documents have been shredded. We might have to sit there and painstakingly put documents together.
The handwriting on a specific document could come into question. Or police would gather handwriting from a specific time and I would go through and methodically look at each letter, how they're linked together, how they've been formed, upper and lower case, punctuation, spelling mistakes. You see if someone's become lazier and started scrawling, the pen they're using, what kind of mood they're in, the surface they're writing on, influence of alcohol, illness. So much can be seen.
There are also indents and impressions. We might get a notebook and see that a page has been torn off. We can still see what's been written through four or five pages. There is the Electrostatic Detection Apparatus ? you put a document on it and it produces a vacuum, drawing the document down. We put what I would liken to clingfilm over the top and run an electric bar over it, which charges it ? the indented impressions will be a different charge. Over this we pour glass beads with a carbon-based powder on them. The powder sticks and reveals the indented impression: words appear, images appear. It's like magic. What you then see is often some sort of malicious communication ? perhaps a threat sent to the prime minister. Something has emerged out of nothing.
In this particular case, there were about 50 very elaborate drawings in pencil and coloured pencil with bits of writing on them which had been obliterated with correction fluid. It was pretty disturbing. There was obviously a purpose for having obliterated those entries.
I was specifically asked to look for women's names, but I wrote down everything I saw: Melissa Halstead's name, Paula Fields' name. I found what I was looking for.