It was a summer night, in July 1964, and Lotfi Zadeh was alone in New York, away from his home in California. He was staying at the home of his parents, who were away, and a dinner he was invited to had been cancelled. Many of us, if we found ourselves in such a situation today, might binge watch that Netflix series everyone had been talking about. This being 1964, and television sets being barely a decade old, Lotfi didn’t have that option. So he sat and thought. “My thoughts turned to the unsharpness of class [category] boundaries,” he later said. “It was at this point that the simple concept of a fuzzy set occurred to me.”
In other words, his thoughts led him to questions. Why, in the world of computer science and engineering, the world in which he was involved, was there so much focus on sharply-delineated categories? It was the heyday of Game Theory, and “models” not tied to the real world ruled the roost in many a department. But weren’t things more ... fuzzy in real life?
This simple idea gave birth to a concept, which Lofti christened Fuzzy Sets and later Fuzzy Logic — a concept that quite simply changed the world. It was first embraced by engineers, who used it in industrial process controls. It played a fundamental role in the design of early smart products, like hand-held camcorders and microwaves. It traversed the globe and found special favor in that tech-nut of a country, Japan. Fuzzy logic was used to help design the underground train system in the city of Sendai in 1987 —a crowning achievement of an idea that had proved itself.
Before Lofti Zadeh passed away on September 7, at the age of 96, he was one of the most influential living minds in computer science and mathematics.
What’s in a Word?
“I could have chosen another term that would have been more ‘respectable’ with less pejorative connotations,” Zadeh once wrote. “[But] I coined the word ‘fuzzy’ because I felt it most accurately described what was going on in the theory. I had thought about ‘soft’, but that really didn't describe accurately what I had in mind. Nor did ‘unsharp,’ ‘blurred’, or ‘elastic’. In the end, I couldn't think of anything more accurate so I settled on ‘fuzzy.’”
That he chose such a term reveals aspects of his personality: serious, yet always a bit relaxed. In 1965, he was far from aware the role this concept would come to play. He submitted his first paper on the concept to the journal Information and Control because he sat on the editorial board. The article was published. Months after, it had only attracted one lukewarm review. Today that article, “Fuzzy Sets,” remains one of the most cited articles in the history of the field. More than 15,000 articles and 12 journals include the term “fuzzy” in their titles. Zadeh’s development of the concept of Fuzzy Logic, in 1973, revolutionized the field further.
During the turbulent 1960s, Zadeh seemed not to care much for all the student activism that made Berkeley a household name worldwide, although he was far from an insular mathematician and remained committed to a generally progressive political vision. In the last year of his life, he was, like most scientists, dismayed by the election of Donald Trump. In the hardworking 1960s, he’d sometimes publish several influential journal papers a year.
In those years, Zadeh was involved in another battle. Should Computer Science be part of the Electoral Engineering department or its own department? Those acquainted with academic affairs know how messy these conflicts can get. In 1967, Zadeh won this battle. Berkeley became home to the Department of Electrical Engineering and Computer Science, a name it still keeps. Universities around the US and the world followed. Far from a simple administrative decision, it changed the way computer scientists developed their craft, and Zadeh would later be presented with a medal for this shift in this field of study, research and innovation.
Baku Birth and Tehran Schooling — But an Internationalist at Heart
Lotfi was, of course, no stranger to conflict. He had been born, in 1921, in a Baku that was just being solidified as the capital of Soviet Azerbaijan after a tortuous civil war. His father, Rahim Aliaskarzadeh, was an Iranian merchant from the Turkic-speaking city of Ardabil who was posted to Baku as a correspondent for the newspaper Iran. His mother, Fanya Korenman, was a Russian Jewish pediatrician from that wonderfully diverse port city, Odessa, and had acquired Iranian citizenship. Lotfi was 10 years old when the family moved back to Tehran, where he studied in the US-founded, Presbyterian-run Alborz high school, a rite of passage for much of the Iranian intellectual elite.
As can be expected with such a transnational figure, nation-states would come to fight over his legacy. The Republic of Azerbaijan claimed him as a leading Azerbaijani scientist, such a big fish clearly useful to a nation that had to build itself in the aftermath of the Soviet collapse. Iranians would, of course, point to his Iranian father, and his studies in Alborz and, later, Tehran University.
Lotfi didn’t seem to pay these fights much mind. He clearly showed a lot of affection for both the Republic of Azerbaijan, where some report he will be buried, and his Iranian past.
"The question really isn't whether I'm American, Russian, Iranian, Azerbaijani, or anything else. I've been shaped by all these people and cultures and I feel quite comfortable among all of them,” he once said. Even a profile in a journal funded by Baku described him as “[belonging] to a world where there are no boundaries limited to time or place... best characterized as an internationalist.”
Still, Lotfi credited his “obstinacy and tenacity” and “not being afraid to get embroiled in controversy” to his Turkic roots. He said without them, he couldn’t have developed Fuzzy Logic.
Scientists have also paid homage to the Soviet civilization that had done its share in shaping his outlook. Not only he had grown up reading the Russian greats such as Dostoyevsky, Chekov and Tolstoy, but he would remember the attention that Soviets had given to science and technology.
“They instilled the belief that you owed something to society,” Lotfi would say decades later. “That you shouldn't be self-centered, egoistic, seeking only your own pleasure. That you should focus on what contribution you could make to others.”
Yet the country that gave Zadeh a chance to develop his ideas was not the Soviet Union, but the United States. He remembered his American teachers at Alborz fondly, people who considered themselves to be genuine “Good Samaritans” who sacrificed so much to help others. It made him want to live in the US. He also hobnobbed with some of the more than 30,000 soldiers that were stationed in Iran during the Second World War, helping his father do business with them.
Even the turbulence of the World War years couldn’t take Lotfi’s focus away. In 1943, he finally found his way to the US. His long name, Lotfi Aliaskerzadeh, was too weird for his adopted country, and so he changed it to Lotfi Asker Zadeh (often written as Lotfi A. Zadeh), which made it a tad strange to both Azeri and Persian speakers instead. He got his Master’s in electrical engineering from MIT in 1946 and then went to New York to join his parents, who had moved to the country as well. He soon got his PhD from the elite Columbia University. Professorship and tenure followed. Lotfi had made it.
The paper he co-authored with his supervisor, John R. Ragazzini, came out in 1952, making him very well known a decade and a half before the Fuzzy Logic years. The paper is often cited even today. But had disputes over the “money-centricity” of the department at Columbia and, in 1959, when invited to go to Berkeley, he accepted. The little sunny town in California became his home for the rest of his life.
In the crucial year of 1965, he went to the Soviet Union for an academic conference aboard the Admiral Ankhimov. He stopped over for a day in Baku. This was his first visit to the city of his birth in 34 years. His third visit would have to wait for 43 more.
A Promoter of “Dignity and Collegiality”
Retirement seemed like a joke to Zadeh, who was actively working well into the tenth decade of his life. He kept corresponding with those interested in his Fuzzy concepts worldwide and must have visited dozens of countries to attend conferences (by 2008, he had racked up 250,000 miles on United Airline flights). He continued to be the subject of many journalists' profiles. Even after his pioneering work of the 1950s and 1960s, he still produced fundamentally new work. One example was a “linguistic variable,” the values of which were were words not numbers — an idea he came up with while watching opera in San Francisco. It led to the important use of natural languages in science and engineering and had it been coined by a lesser-known scholar, it would have been a magnum opus. For Zadeh, it was considered simply one more accomplishment.
Professor Zadeh’s office in the Berkeley Computer Science building continued to be stacked with the latest journals, and he was always happy to welcome students in for a chat. Surrounded as he was by the growing importance of Silicon Valley, he often worried about its work culture. In a commencement speech to Berkeley students in 1997, he called on them to seek not only pay and interesting work but “dignity and collegiality.” He decried the “cut-throat competition” that was defining the culture in which he worked, where "if you make the mistake of stopping for lunch, you will be lunch.”
He did go back to Baku in 2008. By then, of course, it was the capital of an independent Azerbaijan, a country anxious to take credit for the accomplished scientist. He was impressed with the investments President Ilham Aliyev’s government had made in science, and after meeting him he described him as “a truly outstanding personality and a leader with vision and initiative.” Some of his fans must have been disappointed at the praise he bestowed on the autocratic tyrant. But then, the Aliyevs had done much to impress him. He was taken to see Baku’s School Number 16, where he had attended primary school so many years before. More importantly, he also received multi-million- dollar grants from Azerbaijani businesses and the government, both which were happy to fund his research.
The relationship lasted. In 2016, Azerbaijan’s consulate in Los Angeles awarded him a Nizami Ganjavi gold medal, named after a medieval poet that Baku and Tehran also fight over (Nizami hailed from what is today Azerbaijan, but wrote exclusively in Persian). Just a few months ago, in March 2017, a message was published on Zadeh's behalf, congratulating the appointment of Aliyev’s wife, Mehriban Aliyeva, as vice president. He praised her for her work for “human rights and peace.”
Lotfi had fond memories of Iran and was proud of his ability to speak some Persian. When attending a conference in Helsinki in 2008, he recorded a message, albeit in English, for the faculty and students of the Technical College of the University of Tehran, from where he had graduated in 1942. He’d recall that that when he first arrived at MIT, he considered the education in Tehran to have been more rigorous. The cartesian methods his French-inspired professors in Tehran used continued to matter to him — Fuzzy Logic was not about being abstract, he’d say, it was about being precise about imprecise matters. Lotfi remained well known in Iran and Sharif University’s journal, Scientica Iranica, ran a long article about his life in 2011. Many an Iranian student, at home and abroad, named Lotfi as an inspiration.
In the last decade of his life, Zadeh was as active as ever, before suffering a heart attack in December 2008, and further ailments over the last few months. A few months ago, when IranWire was producing a documentary about Berkeley, we tried to interview him but he was unable to take part due to health reasons.
Lofti met his wife Fay in Tehran in the 1940s and they married when they met again in New York in 1946. Born in Japan and having lived in places as diverse as Manchuria, Latvia and Germany, Fay shared a cosmopolitan past with Lofti and was often seen at his side. She also wrote a memoir about their time together. Their two children came to prominence in the US. Stella Zadeh, a talent agency owner and journalist, died tragically in 2006 at the age of 58. Their son Norman Zadeh completed his a doctorate at Berkeley, and has worked for IBM and taught at Stanford University. He also founded an adult magazine — its remit was to only publish photographs of women who had not undergone any cosmetic surgery — and published a book on poker.
Lotfi A. Zadeh is survived by his wife, Fay, and their son, Norman.