Aug 182018

Today is the birthday (1911) of Klára (Klari) Dán, usually called Klára Dán von Neumann, but I want to correct that usage straightaway as part of the problem this post is largely about. Yes, she was married to John von Neumann, and, yes, her married name was Klára Dán von Neumann, but she was a major figure in mathematics and computing in her own right, and deserves to be recognized as such. She was married 4 times. Why should we recognize her by her married name when she was married to von Neumann? Why not use her birth name and recognize her as an individual and not as a woman who gained her identity and importance from the man she married (who also happens to be more famous)? I will call her simply Klára Dán or Dán in this post. Unfortunately, biographical material is in short supply, but I will do my best.

Klára Dán was born in Budapest to Károly Dán and Camila Dán (née Stadler), wealthy Hungarian Jews. Her father served in the Austro-Hungarian Army as an officer during the Great war, and the family moved to Vienna after the war to escape Béla Kun’s short-lived Hungarian Soviet Republic. When the regime was overthrown in August 1919, the family moved back to Budapest.  At 14, Klára Dán became a national champion in figure skating. She attended Veres Pálné Gimnázium in Budapest and graduated in 1929. She married Ferenc Engel in 1931 and Andor Rapoch in 1936.

Klara had previously met John von Neumann (also a Hungarian Jew)  during one of his return trips to Budapest from the U.S. prior to the outbreak of World War II. When von Neumann’s first marriage ended in a divorce, Klára Dán divorced Rapoch, married von Neumann in 1938 and emigrated to the United States. She became head of the Statistical Computing Group at Princeton University in 1943, and moved to Los Alamos National Laboratory in 1946 to program the MANIAC I machine designed by von Neumann and Julian Bigelow.

The MANIAC I (Mathematical Analyzer, Numerical Integrator, and Computer, or, Mathematical Analyzer, Numerator, Integrator, and Computer) was built under the direction of Nicholas Metropolis at the Los Alamos Scientific Laboratory. It was based on the von Neumann architecture developed by John von Neumann. You can see (below) that the architecture is close to modern digital computers with input and output devices, a central processing unit (CPU) and a memory storage device. It did, however have a bottleneck in that input and processing could not be done simultaneously.

It was said that Metropolis (perhaps on the advice of von Neumann), chose the acronym to make fun of silly computer acronyms in the hope they would stop, but it stuck. Early digital computers were hard wired to perform their operations, much like a modern hand-held calculator, although a lot more complicated. To reprogram them took days or weeks to change the wiring and debug the programming. Even booting up was a major operation, so they were always left running. The von Neumann architecture allowed for reprogramming which was where Klára Dán came in. As with all computers of its era, MANIAC I was a one-of-a-kind machine that could not exchange programs with other computers (even other IAS machines). Programmers had to generate unique code for the specific architecture. Klára Dán was the principal programmer of MANIAC I. The first task assigned to MANIAC was to perform more exact and extensive calculations of the thermonuclear process. The MANIAC ran successfully in March 1952 and was shut down on July 15, 1958. It was succeeded by MANIAC II in 1957.

Klára Dán was also involved in the design of new controls for ENIAC (originally used to calculate artillery trajectories) and was one of its primary programmers. Even on the 50th anniversary of ENIAC in 1995, the role of women as programmers went largely unrecognized. All of the first 6 programmers of ENIAC were women: Kay McNulty, Betty Jennings, Betty Snyder, Marlyn Meltzer, Fran Bilas, and Ruth Lichterman. Klára Dán was involved in later modifications. She had a major role in reprogramming ENIAC to make weather predictions. Historians believed for decades that women posing by ENIAC were “refrigerator ladies,” that is, models who made the computer more attractive, but were not involved in its operation. They were not eye candy, they were the key programmers.

After his death in 1957, Klára Dán wrote the preface to John von Neumann’s posthumously published Silliman Lectures, later edited and published by Yale University Press as The Computer and the Brain. She married physicist Carl Eckart in 1958 and moved with him to La Jolla, California. She died in 1963 when she drove from her home in La Jolla to the beach and walked into the surf and drowned. The San Diego coroner’s office listed her death as a suicide.

Maybe some historian or Ph.D. candidate is assiduously researching Klára Dán’s life for a book or dissertation. I don’t know. Her story is largely untold. I don’t have any trouble asserting that John von Neumann was the love of her life, and I expect he valued her for her mathematical abilities, along with other qualities. Why, though, was she married 4 times (the first 2 in quick succession), and why did she remarry soon after von Neumann’s death? Furthermore, why did she commit suicide? If she suffered from clinical depression or was bipolar, her devotion to complex work and suicide would be explicable. But there is so little to go on. She was not necessarily a private person. It is known that she was gregarious in Budapest when her parents entertained, and she and von Neumann held celebrated parties at their house in Princeton. Of course, being overtly gregarious is no sign of anything concerning inner turmoil. If there is more research on her life, I’d like to know about it. Meanwhile spread the word about the undervalued work of women in the early days of computing.

A Hungarian recipe is called for today, and if you think about paprika when you think about Hungarian cuisine, you are not far wrong: You have a hard job finding a Hungarian meat recipe without some paprika in it, and most of them are swimming in it. First task in cooking any Hungarian dish is to find proper Hungarian paprika, not the red powder that passes for paprika in many supermarkets. Then you have to decide on the grade and spiciness. Csípős Csemege, Pikáns (Pungent Exquisite Delicate) is my favorite, but it is hard to find outside of Hungary. You can probably find a hot or mild Szeged paprika if you hunt. You can find good Hungarian paprika online.

Here is Hungarian Pacal Pörkölt to feed my tripe fetish. It is a great favorite of mine. You can skin the tomatoes easily by scalding them briefly in boiling water. To skin the peppers, sear them over a gas burner briefly. Lard is the traditional fat for frying, but you can use olive oil if you wish.

Pacal Pörkölt


2 lb parboiled tripe, cut in thick strips, ½ inch wide, 2 inches long
5 oz smoked bacon, diced
5 oz diced onion
5 oz tomatoes skinned and diced
5 oz Hungarian peppers skinned and sliced in stripsM
3 cloves garlic, peeled and sliced thin
1 tsp ground cumin
1 tbsp Hungarian paprika
1 tsp marjoram
2 bay leaves


Mix the peppers, garlic, cumin and paprika with the tomatoes in a bowl.

Heat a small amount of lard in a deep skillet over medium-high heat and fry the bacon until it is lightly brown. Scoop out the bacon bits with a slotted spoon. The bacon can be set aside or a little cook’s treat. Add the onions and sauté until translucent. Add the tomato mix, ¼ cup of water and the bay leaves. Cover and simmer slowly for about 15 minutes.

Stir in the tripe, marjoram, salt and freshly ground pepper to taste, and ½ cup of water. Cover and simmer until the tripe is tender. Cooking tripe to the right consistency takes experience. It must be al dente – not chewy, not mushy.

Serve hot with csipetke or galuska (Hungarian noodles) on the side. Sour cream can also be served as a side.

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>



This site uses Akismet to reduce spam. Learn how your comment data is processed.