At least as far back as 1500 BC, doctors were aware of an illness which caused excessive urination. Patients with this disease were constantly thirsty because, no matter how much water they drank, it all went straight through them. A Greek doctor known as Aretaeus of Cappadocia invented a name for this disease; his name for it referred indirectly to the disease’s best-recognized symptom. From the Greek diabainein (“passing through”), Aretaeus coined the term diabetes.

If excessive urination had been the only result of diabetes, people would have found a way to live with it, but unfortunately there was more to it than that. The disease also tended to result in a general breakdown of health. It was especially severe when it struck children and adolescents: these young patients could waste away and die within a few months. When the disease appeared in mature adults, it usually took a milder form and it progressed more slowly — but it was still a very serious condition.

What did the doctors of the time do about this disease? Nothing very useful, it seems. In the ancient world, medical practice was a kind of performance art. When Voltaire said that medicine was the art of entertaining the patient while nature cured the disease, he was pretty much on target, at least in terms of pre-modern medicine. The physician’s goal was not to do something useful, but to make his patients think he was doing something useful. It didn't much matter what the treatment was, so long as it wasn’t more dangerous than the disease itself, and so long as the patient could be led to believe that the herbs, lotions, mineral baths, and exorcisms would help. As modern experiments continue to show, patients who assume they are getting an effective treatment (even if they’re really being given capsules full of pastry flour, or injections of water) tend to experience some kind of improvement in their condition. Today we call this the placebo effect. We see it mainly as a problem — a distraction, an obstacle in the path of medical research. In the ancient world, it was not a problem. It was the foundation of medicine.

The placebo effect didn’t have much impact on a condition as serious as uncontrolled diabetes, but the placebo effect was the only anti-diabetes weapon in the medical arsenal of the day, and it would be a mighty long time before anyone invented anything better. Because the real objective of any therapy in those days was to impress the patient, doctors went with their strengths, which had to do with charisma rather than investigative skill. If doctors learned anything at all from experience, it was about how to make a vivid impact on the patient’s imagination. They learned almost nothing about the nature of disease.

What I'm building up to here is that, although doctors have known about diabetes for at least 3500 years, during the first 3300 of those years the only useful discovery they made about it was that the urine produced by people with diabetes was sugary enough to attract ants and bees.

Because of this very odd fact, the name “diabetes” was eventually expanded to “diabetes mellitus”, the latter word referring to the honey-like sweetness of diabetic urine. There is no generally-accepted English translation of the phrase “diabetes mellitus”, but for all practical purposes it means “pissing honey”. (Or, if that is too blunt for you, “passing honey”).

Anyway, the basic idea here is that diabetes mellitus means sugary urine, produced in generous amounts. There happens to be another disease, quite unrelated to diabetes mellitus, which causes the patient to produce excessive amounts of urine which isn’t sugary. This disease is called diabetes insipidus, the second word indicating (rather wistfully, it seems to me) that the urine lacks flavor. Diabetes insipidus is a pituitary disorder, and it has nothing to do with what we usually think of as diabetes, but I thought I’d mention it in case you ever come across the term. Strictly speaking, the word diabetes, by itself, could mean any disease that can cause excessive urination. As a practical matter, though, when the word diabetes appears by itself, you are pretty safe in assuming that diabetes mellitus is meant by it.

For ancient doctors, the presence of sugar in the urine of their patients — in sufficient concentrations to attract swarms of insects — was only a quirk which made diabetes mellitus easy to recognize. From a modern perspective, the fact that patients were expelling sugar from their bodies by way of the urinary tract is an obvious clue to the nature of their illness, but ancient doctors had no idea how to interpret it. All they knew about the disease was that it made you produce a lot of sugary urine (and then made you die). Because they had no insight into the nature of the disease, doctors had no idea how to treat it. They tried all sorts of things, though. Most of what they tried was medically useless. One Greek doctor recommended horseback riding; at least that’s a kind of exercise. Generally speaking, though, doctors could make no progress in treating diabetes mellitus, and there the matter rested for over three thousand years.

Eventually the scientific revolution came along, and useful investigations of diabetes began to get under way. In 1798, it was shown that diabetic patients had excess sugar not only in their urine, but in their blood as well. In fact, the sugary urine had only been a secondary effect. The reason these patients were producing so much urine in the first place was that their blood was full of sugar, and their kidneys were working overtime to get rid of it (by extracting the excess sugar from the bloodstream and dumping it into the bladder.) The body lost a lot of water in the process of flushing all that sugar out of the bloodstream, which is why diabetes patients tended to be chronically thirsty.

Once diabetes mellitus began to be defined as chronic excess sugar in the blood, and obvious question arose: why should sugary blood be a problem?

Obviously, excess blood sugar was harmful in some way; this was suggested by the fact that the kidneys were working so desperately hard to get rid of it, and confirmed by the fact that, when the kidneys couldn’t get rid of it fast enough (so that it built up to extremely high levels in the blood), the patient’s health began to fall apart. But exactly why excess  blood sugar was harmful was not clear. Nor was it clear why diabetics had so much more of it than other people. They didn’t seem to be eating more sugar than other people. Apparently the body had some kind of mechanism for regulating the concentration of sugar in the blood. Sometimes that mechanism broke down, and the patient became diabetic. But how could the mechanism break down?

In 1889 another important discovery was made. Experiments designed to uncover the purpose of the pancreas (through the rather stern method of removing it from animals, and waiting to see what happened to them) revealed that any animal that lost its pancreas immediately developed severe diabetes mellitus. Apparently the pancreas played a critical role in regulating sugar levels in the bloodstream. In some people, the pancreas wasn’t functioning normally, so the sugar level got out of control. (By the way, there are various forms of sugar, and the form that is found in blood is called glucose; I will mostly use that term from here on.)

Doctors began to make extracts of pancreatic tissues and inject them into diabetes patients. It sounds crude. It was crude. But it did help some patients, which seemed amazing at the time.

Eventually it became clear that the useful ingredient in pancreatic tissue was a hormone. The hormone was produced by small, isolated patches of tissue in the pancreas. These patches were known as the “islets of Langerhans” (after Paul Langerhans, who had first described them in 1869). Because the hormone came from these “islets” — and the Latin word for islets was “insula” — the hormone was named “insulin”.

It turned out that the function of insulin was to stimulate body tissues to start absorbing glucose from the blood. Most body tissues wouldn’t absorb glucose under ordinary circumstances — they had to be commanded to do so, by a release of insulin into the bloodstream. The more insulin was released into the bloodstream, the faster the tissues absorbed whatever glucose was present in the blood. If no insulin was released into the bloodstream, the tissues absorbed little or no glucose, no matter how much glucose was present in the blood.

Apparently the explanation for diabetes was that, in some people, the islets of Langerhans in the pancreas had stopped producing insulin. Without insulin, glucose wasn’t absorbed into the tissues. The glucose had nowhere else to go, so it built up in the bloodstream, until the blood became so sugary that the the kidneys started flushing glucose into the bladder. That was the point at which the patient started to develop a fan-base among the bees.

No one knew why the pancreas would sometimes fail in this way, but it was clear that. when such a failure occurred, the result was a sudden onset of severe diabetes. That, of course, is exactly what happened to an animal when its pancreas was experimentally removed, but apparently the same thing could happen to a person with an intact pancreas — because the organ had, for whatever reason, stopped working.

In 1921, doctors developed techniques of extracting and purifying insulin from pancreatic tissue, so that it could be made into an injectable medication. When diabetes patients were given insulin shots, their glucose levels declined immediately. In fact, there was some danger of over-correcting, and driving the glucose level dangerously low. Doctors had to be careful not to give patients too much insulin at once.

There were a few other practical hitches. For one thing, insulin was broken down in the process of being used by the body, which meant that the glucose-lowering effect of insulin injections was temporary. You had to keep taking insulin shots to prevent your blood glucose from building up again. Also, insulin would be destroyed by digestion, so it wasn’t possible to take insulin orally. If you wanted to take insulin, you had no choice but to bypass the digestive tract and inject it directly into the bloodstream. These inconveniences were a small price to pay, however. Diabetes patients whose glucose levels were maintained at a normal level by means of regular insulin shots showed a remarkable improvement in their health. Diabetes, formerly a mysterious and unstoppable killer, had become a manageable chronic condition.

However, there were some oddities that still needed to be explained. Surprisingly, the insulin treatment worked better for patients with the more severe “juvenile” form of the disease than for patients with the less severe “geriatric” form. You’d think it would be the other way around, wouldn’t you? And yet, people who became diabetic later in life (and whose diabetes was usually less severe) seemed not to get as much benefit from insulin, unless it was provided in large doses — sometimes in a dose so large that it would have endangered the life of a younger patient. Clearly, patients with late-onset diabetes had a slightly different problem than patients who developed the disease in their youth.

Resarch on this confusing problem continued, and in the 1950s a reasonably clear answer began to emerge. It turned out that the patients with the “juvenile” type of diabetes were the ones who had lost their ability to produce insulin. The patients with the “geriatric” type of diabetes could still produce insulin, but they were becoming less and less able to respond to it. Their body tissues had become insensitive to insulin (or “insulin-resistant”, as the problem came to be described). For an insulin-resistant patient, being able to produce ordinary amounts of the insulin was not good enough. Only an abnormally large amount of insulin would work for them, and although they were able to produce insulin, they weren’t able to produce enough extra insulin to overcome their insensitivity to the stuff. That is why, when they were treated with insulin shots, they often required a bigger dose than would be needed for a patient with “juvenile” diabetes.

Explaining the differences between type 1 and type 2 diabetes (lack of insulin vs. insensitivity to insulin) was helpful, but it left many questions unanswered. What exactly was causing these insulin-related disorders? Why were some people losing their ability to produce insulin? Why were other people retaining the ability to produce insulin, but losing the ability to respond properly to it? Could anything be done about these problems besides taking insulin shots? And why did it matter in the first place — that is, why was excessive blood sugar, regardless of its cause, so harmful to the patient’s health? I will discuss these questions in other essays here.