We frame our expectations and our insights by the stories we tell. This is true as well of the stories of innovation. How does something new come into the world and reshape things? We have two primary narratives–bane and boon–one, in which a bane is turned away by the status quo, and the other in which a boon is found or made to advantage. Variations a numerous. Banes from the outside, like Grendel tearing up Hrothgar’s hall, and banes from within, like heresies and seditions ruining a good thing. Boons of discovery, like those of sea voyages and research, and boons of cleverness, by which a new thing is made. And we have warnings for both. How the things we fear may be for our benefit, or how what we make can go bad and even detest us for it (Caliban, Frankenstein’s monster).
If we look at bane and boon in innovation narratives we have this same geography. We have innovation that succeeds (Edison, Sarnoff) and innovation that gets beaten back (Tesla, Farnsworth). If there is going to be a boon narrative, however, we expect that it has to end up with “and that’s how we got to this thing today.” It ends up with the innovation in practice. The ending is clear. The thing that matters is where we decide to start the narrative, what details to choose, and most importantly what we say about how things have got to this necessary end point that we call out as success.
What matters in the construction of narrative? One is narrative itself. Telling a story has a particular dynamic to it. There is action and choice, crisis and resolution. The thing has to look like a story, not like a table of data or a list of facts related only by topic. But there is also the more mysterious thing, the premise of “what actually happened.” This itself requires artiface–that it is possible that some one thing “actually happened” despite the “welter” of bits and pieces of event, intention, half-action, serendipity, cause and effect. What did “actually” happen? That’s not merely a matter of “facts.” It’s a matter of context, of observers and participants, of states of mind and later reasons, of random things that look connected and connected things that appear random, of multiple intentions and opportunities that compete and confuse and conflate and converge. A narrative not only picks and chooses, but creates and suppresses. There is no other way but to navigate the welter of it.
Finally, there is coherence. A narrative has an action. It obeys conventions of time and place and causality. It moves from beginning to middle to end. It has closure. It’s hard not to leave off with an implied “and that is how it is today.” One can leave off with “you thought you knew, but now you really know.” Or “you knew, but you didn’t know why.” But it’s really all of a kind, and not “and that’s how it isn’t today, and never was.” This need for a coherent action in narrative is one of the biggest challenges in getting things right, to hold in place something worth remembering, and remembering it as if experienced, not merely as a fact or sounds washing over us. Coherence in narrative, one might say, introduces artifacts that patch together pieces selected for telling. Narrative in this sense rationalizes the welter of the world. We cannot help it. The issue is not how to get around such artifact, but what quality of rationalizing are we going to do, and what does that lead to by way of our own next actions? Does life imitate art? If so, what art are we experiencing, and what art are we creating to experience later?
We have bane and boon. We have worries about both. And we have the composed narrative and the assumed welter. And we have rationalizing artifacts of narrative. With all this in mind, let’s look at some narratives around an innovation. How about warfarin? Warfarin is a blood thinner developed at the University of Wisconsin. (There, that’s a simple boon narrative, no worries). It is also one of the founding innovations of the effort behind university research in the public interest. One might connect it to the Cottrell invention and formation of Research Corporation, or more locally to the formation of the Wisconsin Alumni Research Foundation in the 1920s, or to the Morrill Act and the more local Wisconsin Idea, under which no aspect of community life is beneath the dignity of university research to understand and seek to improve and empower.
Here is a simple account of warfarin:
History of Warfarin
It can’t be denied, warfarin is used as rat poison. But that’s like saying nitroglycerine is used to explode boulders. While true, that has little to do with using nitroglycerine under the tongue to relieve angina pain. Similarly, warfarin as medication, when used correctly, is not dangerous and can be life-saving.
Warfarin has a background in agriculture. It’s found in grasses and grain fields. Cattle that had excess bleeding from eating moldy hay led to its discovery. The smell of freshly mowed grass is largely from coumarins, compounds related to warfarin.
In the 1950s, the medicinal use of warfarin in humans was developed. As a medicine, it is warfarin sodium, a crystalline powder that is highly soluble in water. Trade names besides Coumadin are Jantoven, Marevan, Lawarin, and Waran.
Some cleverness, but about as generic as one can get. Warfarin “was developed”. Reduced to Piscopo (one of my favorite languages): Bleeding cows! Rat poison! Medicine! Products!
A similar simple version is here, under the heading of “world-changing drugs discovered by accident”:
Known as Warfarin and also by it’s brand name of Coumadin, the blood thinning agent has had an interesting history to say the least. It was discovered when farmers realised their cattle were dying from eating a type of clover and they found out an active chemical in the plant was preventing the cattle’s blood from clotting, causing them to hemorrhage.
They then decided to use the active chemical as a rat poison as it was an easy way of killing rats. Then later on, after more research they realised that in correct doses it proved beneficial to people who had blood clotting problems and in 1952 Warfarin was first used as a blood thinning agent for humans.
More interesting. But as we will see, it gets things wrong (the agent was not in the plant but in a by-product of a fungus acting on a molecule in the plant), and it was discovered a long time after the farmers had figured out the problem (hay going bad) and the solution (change feed immediately). What got discovered didn’t work so well for rats, so lots of other work had to be done to develop something related. And we will see that something rather shocking is covered over by “…after more research, they realized….”
WARF provides what we might call an “official” account. WARF is justifiably proud of its role in developing warfarin and related products. At its web site, we find this account:
Coumadin, and its counterpart, Warfarin, together represent one of the first technology transfer success stories emanating from the University of Wisconsin-Madison’s Wisconsin Alumni Research Foundation (WARF), UW-Madison’s patenting and licensing arm, which has supported the university’s scientific research since its establishment in 1925.
The story begins in 1933, when a farmer from Deer Park, Wis., paid an unexpected visit to Professor Karl Paul Link’s laboratory in UWMadison’s School of Agriculture. The farmer’s cows had been dying, and he suspected it had something to do with the sweet clover hay the cows had been eating. For that reason, the farmer had brought samples of the clover feed and container of non-coagulated blood from one of his cows to Link’s lab.
In 1941, after years of research, Link and his team isolated the anticoagulant in the clover feed. The researchers found that it was highly toxic for rodents and eventually patented it under the name of Warfarin (named after WARF), for use as a rat poison. It ultimately became one of the most widely used rat poisons in the world.
Further research on Warfarin yielded several related compounds, which also were patented and used in medical practice. Coumadin®, a blood thinner for treating heart patients and preventing blood clotting, was among these compounds. In the years since, Coumadin has become the most widely prescribed blood thinner in the world.
This account provides new elements. The context is technology transfer success stories. Not only was there research, but this was university research, and it later resulted in commercial products–Coumadin and Warfarin. The story brings us a starting point, a vist by a farmer. “Unexpected” covers for a welter of details, which I will get to in a bit, and you will too if you keep reading. Professor Karl Paul Link wasn’t a full professor at the time–he had just arrived at the University of Wisconsin himself, and the farmer (not named here) equally had not expected to show up at Prof Link’s lab, as he had intended to show up elsewhere (and had). We see that it took eight years for the lab to figure out what was causing the cows to bleed to death (but we don’t see that the lab changed its research direction as a result of the visit, which it did, apparently–but it was of no help to the farmer, who had dying cows to deal with, not eight years to wait). We also see that researchers “found” it was a good rat poison (more of that later), and with further research (again suppressing shocking details) it also “yielded” medical uses. These two uses are both marked as the “most widely” used “in the world.” Heady stuff. It is a success story. It ought to end up that way. It is necessary once one labels something for success.
Accounts such as “A Short History and Useful Factoids about Warfarin (Coumadin)” and this one, at Wikipedia, repeat the start in the problems of farmers with dying livestock, introduce Professor Link. “Short History” has it:
A researcher by the name of Karl Paul Link, working under the aegis of the Wisconsin Alumni Research Fund (WARF), did a careful analysis of the ensilage from ranches that suffered losses and those that did not. He discovered that a chemical, dicoumorin, found in the ensilage of sweetclover hay from those ranches suffering the losses, was a powerful anticoagulant. Dicoumarin is the result of a substance called coumarin, which is the chemical which gives new-mown hay its characteristic smell, being subjected to the heat and mold in a silo, and forming a double molecule. The year of the serious losses had been an unusually warm one after the ensilage was created. [This information is from Kingsley’s book “Poisonous Plants”, which is one of the first serious studies of the biology of plant toxins].
Here we get attribution to WARF for providing the “aegis” (later, we will see there was WARF funding, but not necessarily right off), we don’t get the farmer story, but we do get an account of why the hay was getting moldy. We also get some insight into manufacturing issues:
One of the problems of warfarin is the process that manufactures it tends to leave a number of deleterious impurities in the resulting product. When used as a rodenticide, this is not a major consideration. And these are only dangerous over a sustained period of time. But it was also determined that a good anticoagulant for human beings would be a Good Thing. The product trademarked Coumadin is a form of crystalline sodium warfarin that is created by a process that leaves no such impurities, and therefore is safe for long-term human consumption.
“Short History” touches on an important event as one of its factoids:
Warfarin was used in the 1950s as an anticoagulant for victims of heart attacks and strokes, but gained fame when it was used to treat President Dwight D. Eisenhower after his 1956 coronary (while in office).
It ends with an account of how warfarin works in therapeutic situations and the problems of those on warfarin therapy, including a personal account of when pigging out on shrimp may not be a good idea.
In the Wikipedia article (in the form I have it presently–it could change at any time), after spending a paragraph showing that the cattle dying problem started in the 1920s and moldy hay was the problem, provides this account:
The identity of the anticoagulant substance in spoiled sweet clover remained a mystery until 1940. In 1933 Karl Paul Link and his lab of chemists working at the University of Wisconsin set out to isolate and characterize the hemorrhagic agent from the spoiled hay. It took five years for Link’s student Harold A. Campbell to recover 6 mg of crystalline anticoagulant. Next, Link’s student Mark A. Stahmann took over the project and initiated a large-scale extraction, isolating 1.8 g of recrystallized anticoagulant in about 4 months. This was enough material for Stahmann and Charles F. Huebner to check their results against Campbell’s and to thoroughly characterize the compound. Through degradation experiments they established that the anticoagulant was 3,3′-methylenebis-(4-hydroxycoumarin), which they later named dicoumarol. They confirmed their results by synthesizing dicumarol and proving in 1940 that it was identical to the naturally occurring agent.
Importantly, here we see that Professor Link worked with a team, some students, to get the work done. The account goes on to discuss the difference between what was found in plants and the chemical form of warfarin. We find out that dicoumarol was patented in 1941 and that warfarin is first used as a rat poison in 1948. We find out as well: “Although warfarin was developed by Link, the WARF financially supported the research and was assigned the patent” (citing the 1959 account by Professor Link which we will get to soon enough). We also see in this account how things moved into the therapeutic use:
After an incident in 1951, where a US Army inductee unsuccessfully attempted suicide with multiple doses of warfarin in rodenticide and recovered fully after presenting to a hospital, and being treated with vitamin K (by then known as a specific antidote), studies began in the use of warfarin as a therapeutic anticoagulant. It was found to be generally superior to dicoumarol, and in 1954 was approved for medical use in humans. A famous early recipient of warfarin was US president Dwight Eisenhower, who was prescribed the drug after having a heart attack in 1955.
The original therapeutic dicoumarol was not, apparently, all that effective. It is with the suicide attempt that things come into some relief. After overdosing on warfarin, the guy still lives. This is the event that turns interest and puts a counterpoint on the president, commander and chief and former general, using warfarin following a heart attack. This is good stuff for stories.
We get a further point: “The exact mechanism of action remained unknown until it was demonstrated, in 1978, that warfarin inhibits the enzyme epoxide reductase and hence interferes with vitamin K metabolism.” The research that led to the development of the commercial product did not produce the science behind the effect. The science came 45 years after the farmer shows up at the lab with a milk can of cow’s blood. What are we to make of what the lab accomplished before the science was complete? We see that there were observations, there was identification and characterization of substances, and selection of these substances for their desired effect. We see that this work was done in a university lab, and was funded by WARF. But it takes half a century for the science to establish why the substance does its thing. We might see from this that there are three branches to the science–first, there is a lab that can receive farmers with tubs of blood (and other stuff, like silage and dead cows); second, that the lab can do studies involving biochemistry to observe and develop compounds with interesting properties; and third, that the lab can develop new compounds that are introduced as commercial products.
We note that one thing that that the lab does not do is solve the farmer’s problem or develop something to make moldy hay safe or to remedy cattle eating moldy hay or to provide hay with properties so it doesn’t get so moldy, or at least doesn’t develop a bad kind of mold, or any sort of early detection system for hay mold or cattle bleeding–no red cow alert–nor any way of storing hay so that mold is less likely to form. No, none of this. The lab goes off in a different direction, following one line of science and not others, and produces a couple of widely used commercial products. The science that shows how the compounds operate comes later, from others. The working knowledge that deals with dying cows and moldy hay comes, as well, from others.
A different take on all this comes from “Warfarin and Chinese Medicine,” which doesn’t have much to do with Chinese medicine but does set some interesting context for warfarin. In “Chinese Medicine” we start with dairy farming in the midwest, we look at the growth of agricultural chemistry, and we get an account of Harry Steenbock’s developments in the use of radiation to enhance the vitamin D content of milk without adding anything to the milk, thus working around regulations that did not permit milk to have additives. Nifty use of science. It’s Steenbock’s workaround that gets a patent and commercial success and leads to the formation of WARF in 1925.
We then get the account of problems with hay. We find that sweet clover is from Europe, holds more water, and therefore is more susceptible to mold. We also find out that two other University of Wisconsin researchers were trying to develop a strain of sweet clover that was more rot resistant, and that this was what Prof. Link was working on when the farmer, Ed Carlson, shows up. In this account, we have the name of the farmer. This is of some import. Names draw attention to who matters. Prof. Link clearly matters, since he is identified as the inventor and primary mover behind commercial efforts. “. . . Carlson’s visit stimulated intensive follow-up.” Well, it appears that it changed the whole direction of the lab’s efforts. We get some further details on plant enzymes, and then this comment on the commercial efforts:
Dicoumarol was patented as an anticoagulant in 1941. However, its action was difficult to control, the dosage range was very large (25-200 mg/day), and it was never extensively used, partly because another anticoagulant was of greater interest–heparin (isolated from cattle liver and lung, and purified sufficiently for human use in 1937). Heparin was relatively safe to use, but had to be given intravenously. It remains a valuable therapy to this day, with a modified low molecular weight version, enoxaparin, introduced recently.
Link continued his work with coumarin and dicoumarol. But, he turned to the problem of rodent control, which was another issue faced at dairy farms (and elsewhere). Since his work with dicoumarol was traced to the problem of hemorrhaging in cattle, he thought of using the compound as a rat poison, to cause hemorrhaging and death. It turned out to be too weak and unreliable for that purpose, so he and his coworkers synthesized some variants and successfully isolated a more potent compound. Patent rights were assigned to the foundation that funded the work, WARF, from which the name warfarin was derived (WARF + coumarin). In 1948, warfarin was launched as a new commercial product: the ideal rat poison. It is currently the most widely used rodenticide in the world.
We get to the success story end point, but the pathway adds some complications. First, dicourmarol didn’t work out. Difficult to control! Large doses! Not used much! Heparin!
Heparin–a competing product discovered at another university (Johns Hopkins) (er, make that, a medical student discovers something, his faculty supervisor figures out another compound and calls that by the same name) developed by another university (University of Toronto)(er, Connaught Laboratories, “an uniquely organized, non-commercial and self-sustaining part of the University of Toronto from 1914 until 1972). Drat and double drat. So Link turned to killing rodents, which took some real effort. The account here breezes over how Link came to think of killing rodents, but no matter, we will get around to that, too.
In addition, we get the story of the attempted suicide:
Tur[n]ing a rat poison into a widely used drug for humans may seem odd, and its path to success was paved by an unusual series of events. In 1951, a navy recruit unsuccessfully attempted suicide by taking rat poison. He had consumed over 500 mg of warfarin (100 times the standard clinical dose of the drug as now used in humans), seemingly more than enough of a rat poison to cause death. Yet, he made a full recovery. This led to research into the potential of using warfarin instead of dicoumarol in humans, since it was a more potent anticoagulant, yet not acutely toxic. In fact, early testing showed that it was far superior to dicoumarol, and clinicians soon discarded dicoumarol in favor of warfarin. The rat poison turned drug was introduced commercially as a human anticoagulant in 1954. It was soon put to very good use: President Eisenhower was treated with warfarin following a heart attack during his early presidency, in 1955, and it is credited, along with other medical interventions, with saving him. He remained president until the end of his second term (January 1961) and lived until 1969 (reaching the age of 79).
It is an odd turning point, that a rat poison made from something that didn’t work as a rat poison turns out to be a better therapeutic than the version that was on the market as a therapeutic. We have from 1933 to 1954 to get to a therapeutic that works decently, six years after the development of the compound as rat-bane. Our account then gets to how warfarin works and Chinese herbs involved in vitalizing the blood (salvia looks good) and looks ahead to future therapies, with hope that there will always be a role for Chinese blood-vitalizing herbs.
If we look at a retrospective by one of Link’s students, we can fill in some details of the rat poison origin. Helpfully, we get two alternative narratives from the welter. First:
During a six-month “sabbatical” at a local sanatorium in 1945, while recovering from “wet pleurisy,” Link conceived of the use of a coumarin derivative as a rat poison. During this period of enforced idleness, he reviewed all of the chemical and bioassay data from his laboratory and selected candidate relatives of dicoumarol that had been synthesized between 1940 and 1944, as potential “better mousetraps (rat traps).” A more potent analogue of dicoumarol named warfarin was first promoted in 1948 as a rodenticide.
From this we get a leading researcher for health reasons away from the lab, getting an idea, reviewing the literature, and going after a practical application. That makes some sense. But the alternative narrative also makes some good sense:
In another version of this story, Link was extremely concerned by a 1948 report from England that dicoumarol might be useful as a rat poison. He felt that such a development would doom the use of dicoumarol as an anticoagulant, as clinicians would be reluctant to prescribe “rat poison” to patients. He set out to find an analogue from the lab’s collection of previously synthesized relatives of dicoumarol, and selected Warfarin (compound #42, a coumarin derivative) for this purpose. It is ironic that Warfarin (Coumadin) rapidly also became the drug of choice for clinical use, largely replacing dicoumarol, and achieved perhaps its greatest fame (notoriety) when used to treat then President Eisenhower after his heart attack in 1955.
Now we have two competing accounts, both of which have value. In one, we think, well, perhaps it would be good for researchers to be required to step away from their labs and funding and research and all that. They might think of something more immediate and practical. In the other, we think, well, perhaps it would be good for researchers to take a more pro-active role by using their labs to defend their previous efforts at commercialization–not only might this help, but hey it looks like one might come up with something even better than the stuff companies holding a license are trying to sell already. Either way, there’s something to consider, and it’s deeper than “then those researchers did this, and then that, and then, world success.”
Perhaps from all of this we have enough to deal with an account such as “The Story of Coumadin.” This account, by Jeff Guillory, starts out with a poison theme, touches on rat poison, and gets into cow killing times: “The cattle ranchers were at a complete loss to know why this was happening. Veterinarians and scientist began to respond to pleas from the ranchers for an answer.” The poor farmers didn’t have clue. Only credentialed professionals might save them. We get the story of Ed Carlson, our farmer:
One day in the Winter of 1933 a farmer named Ed Carlson, drove to the biochemistry building at the University of Wisconsin with a dead cow, a bucket of un-clotted blood and a truckload of old sweet clover that had been cut and used for hay. He pleaded with the researchers to help him with his dying cattle. Professor Karl-Paul Link did research at the university and began to work on this project.
As we will see from Link’s account, Carlson did drive to the lab, and he did plead, but the project that Link worked on was not the one that would help Carlson at all. Link worked out some science, and then worked on ways to exploit the effect rather than suppress it. So Link didn’t help the farmers at all. We then get the first story of insight:
Later in Professor Link’s career he took a sabbatical from his research because he had tuberculosis and spent time in a sanitarium. While there he read about a troublesome problem with rodent control and thought of a use for the “coumarin” which his team at the university had discovered. Professor Link patented warfarin, which he developed into rat poison.
Which leads us to the suicide attempt, with the idea that doctors thought up the idea of administrating vitamin K to save the man’s life so “now medical researchers knew how to counter-act an overdose.” Following this is a passage regarding research to apply warfarin therapy to stroke. We get this piece of the story:
Coumadin (warfarin) was not used frequently until 1955 when President Dwight D. Eisenhower had a heart attack while on vacation at his in-laws’ house in Denver. Coumadin was used to prevent clots from forming in President Eisenhower’s damaged and weakened heart. Eisenhower’s long term treatment included Coumadin 35 mg per week. Since then, it has been one of the most widely prescribed drugs in the United States.
It’s a nice set of transitions from the suicide attempt to medical research to a timely need to save the president’s life in Denver, and from that the drug goes on to its wide use and success.
Next up, let’s look at Professor Link’s account, from 1959, when warfarin had established itself in two major commercial markets, as rat-bane and as presidential-boon.