Memorial Tributes: Volume 27
IsNew No
Tribute Author
Membership Directory

Search this Publication

  • HERMANN A. HAUS (1925-2003)



    HERMANN ANTON HAUS died unexpectedly on May 21, 2003, at the age of 77. He was born Aug. 8, 1925, in Ljubljana, which was then in Yugoslavia. His early education focused on languages and the classics; but he spent much of his spare time as a youth designing and flying model gliders. At age 18, just as he was about to enter the university to begin studies in engineering, the Germans invaded the country and the university was closed. To avoid the draft, he became an electrician trainee in a local aluminum factory and worked there until after the war.

    During that period, several events occurred that would play important roles in his future. Hermann lived alone with his mother, Helene Hynek, who had separated from his father, Otto Maximilian Haus, a noted physician and tuberculosis specialist. One evening a woman partisan who had escaped from a German prison appeared at their house and asked to be hidden. In spite of the risk they helped her, and the next day she was able to flee safely to the countryside. A year later, after liberation from the Germans by Tito’s communists, Hermann and his mother were jailed as possible dissenters. Two weeks later, without explanation, they were released. It seems that the woman partisan they had protected had been made minister of education in the new government, and she arranged to free them.

    Hermann returned to the factory and a warm welcome from his coworkers, who had come to like and trust him. He was even elected company overseer of their adherence to communism because they knew he wouldn’t betray them. This period of freedom for Hermann, however, lasted less than nine months. One night, without warning or time to grab belongings, he and his mother were hauled out of their house by the communist authorities, put on a train in an unheated cattle car, and deported to Austria. One of the other passengers on the train, a chemistry professor that Hermann recognized, lamented having to leave behind a career’s worth of research notes. Hermann vowed to himself never to rely on anything so complex that he couldn’t keep it in his head. This vow greatly affected his career: He was adamant about giving lectures without notes, and in his research he often created elegant derivations of well-known results as well as the completely original and foundational formulations for which he became renowned.

    After arriving in Austria, Hermann and his mother lived in two refugee camps in succession until he secured a job as an electrician with the British occupying forces. He knew some English from reading Gone with the Wind. They moved to Graz, and in 1946 Hermann enrolled in the Technical University there. He chose power engineering in order to learn the principles of the electric motors that he had worked with in the factory. This gave him his first encounter with Maxwell’s equations, which provided the foundation for much of his future work.

    At his high school (gymnasium) in Ljubljana, Hermann had learned Latin and Greek but no calculus. He had some catching up to do. Nevertheless, he quickly distinguished himself and, after discovering an error in a well-known electromagnetics text, learned not to believe everything written in a book. He had determined for himself that a statement in the text, “The curl of a vector is always perpendicular to that vector,” was not rigorously true. Emboldened to pursue more modern applications of Maxwell’s equations, he transferred to the Technical University of Vienna to study microwave engineering.

    In Vienna he heard about a scholarship program for study in the United States and applied. But after waiting for what seemed to him too long without a response, he wrote to U.S. Army General Mark Clark and suggested that Austrians seemed to be discriminated against in the granting of scholarships.

    Whether that letter turned the tide or not, Hermann soon received the offer of a scholarship to study at Union College in New York. With credit for his studies in Graz and Vienna, he graduated after one year (in 1949) with a B.Sc. in electrical engineering.

    He applied to Bell Laboratories for a job and was turned down. He applied to MIT for graduate school and was turned down. He applied to Rensselaer Polytechnic Institute and was accepted and given a teaching assistantship.

    While working on his master’s degree at Rensselaer Polytechnic Institute, he got a summer job at MIT with Louis Smullin (NAE 1970). That experience led to his acceptance in 1951 into the doctoral program of MIT’s Department of Electrical Engineering. Although he continued to work with Smullin on traveling wave tube (TWT) experiments, it was his thesis advisor, Lan Jen Chu, who mentored him in TWT theory and, Hermann later said, taught him the value of simplifying a problem to its bare bones without losing its essential features. His thesis proved that the noise output of a TWT had a lower limit proportional to (G - 1), where G is the power gain;1 he received his Sc.D. from MIT in 1954 and joined what is now its Department of Electrical Engineering and Computer Science.

    The evaluation of noise and the limits it imposes was a theme throughout Hermann’s career. In work with Richard Adler, he extended his analysis to linear electronic amplifiers in general. The two researchers showed that every linear amplifier could be characterized by an optimum noise measure, M, the lowest value of which was (F - 1)/(1 - 1/G), where F - 1 is the excess noise figure.2 Intrigued by the fact that there was no classical lower limit to M, Hermann pushed further and, with James Mullen, found the quantum limit: They revealed that the amplified spontaneous emission power in a single-mode amplifier had a minimum value of (G - 1)hω0B.Elegant quantum noise experiments with Charles Freed then followed. With an early helium-neon laser, Haus and Freed proved the theoretical prediction that photon statistics changed at laser threshold — from degenerate Bose-Einstein below the threshold to Poissonian above.

    During this period, Hermann was also advancing his mastery of electromagnetic theory. Chu had recently extended Wolfgang K.H. Panofsky (NAS 1954) and Melba Phillips’ theory of moving polarized media to moving magnetized media by modeling the magnetic dipole, in analogy with the electric dipole, as two magnetic charges separated by a small distance. This approach was challenged in a paper by Bernard Tellegen, who showed that the force on two magnetic charges in a time-varying electric field was not the same as that on a current loop — the conventional physical model for a magnetic dipole.

    Hermann and one of his former students, Paul Penfield (NAE 1994), took up the challenge. They showed that Tellegen had erred by assuming that the distribution in the current loop was unaffected by the varying electric field. By including the correct induced-charge accumulations, they proved that Chu’s simple formulation was in fact relativistically self-consistent and led to the correct answers. Empowered by this result, they developed the theory of electrodynamics of moving media further and wrote a definitive book on the subject, Electrodynamics of Moving Media (MIT Press, 1967).

    In 1974-75 Hermann spent the first of three very productive sabbaticals at Bell Labs in Holmdel, New Jersey. At that time, two emerging technologies attracted his interest: ultrashort-pulse lasers and integrated optics. Femtosecond-duration pulses had recently been produced for the first time, using a passively mode-locked dye laser. Hermann set about developing a theory for how such short pulses were produced in a laser, given the fact that none of its elements could respond on such a fast timescale.

    In 1975 Hermann provided an elegant analytic theory that could be used to determine stability criteria, optimum operating conditions, and output characteristics. This theory of mode locking with a “slow saturable absorber” became one of his most widely acclaimed works.5 For comparison and completeness, he then formulated an equally elegant analytic theory for mode locking with a “fast saturable absorber.”At the time such an absorber was merely a mathematical construct; but it was prescient. With the discovery some 15 years later of artificial fast absorbers made possible by reactive nonlinearities in solid-state and fiber lasers, this paper joined Hermann’s most highly valued and used works.

    During the same sabbatical Hermann was also attracted to the concept of lasing in a periodic structure, as put forth by Herwig Kogelnik (NAE 1978, NAS 1994) and Charles Shank (NAE 1983, NAS 1984). The distributed feedback (DFB) laser in its simplest form exhibited two equally likely lasing frequencies. Hermann very quickly devised a solution to this uncertainty and, with Shank, published a paper describing the concept of a DFB structure with a quarter-wave shift in the middle.7 DFB structures in semiconductor diode lasers were subsequently widely used as one of the most important components of optical-fiber communication systems; and they remain critical devices for achieving the wavelength precision needed in advanced high-capacity systems. Ultimately, however, Hermann’s quarter-wave shift, while an elegant and effective invention, turned out to be difficult to manufacture, and other methods were found to achieve the same result.

    Upon his return to MIT, Hermann put several of his students to work on ideas for mode locking and integrated optics that he had developed at Bell. No one had mode locked a semiconductor diode laser, but Hermann saw similarities between these devices and the dye lasers he had recently studied. Not having the fabrication facilities to make a semiconductor device that incorporated both a saturable absorber and a gain section, he opted for active mode locking, which was induced with an applied electrical modulation. By 1978 he and his student Ping Ho had produced picosecond pulses with the first mode-locked semiconductor laser, triggering several decades of widespread research on semiconductor mode-locked lasers for high bit-rate optical communications and clocking.

    At the same time, Hermann was developing ideas for optical signal processing in integrated optics. He focused on the design of a switch in which one (control) optical pulse changed the path of another (signal) optical pulse, at picosecond speeds and without changing the wavelength of the switched signal pulse. In 1983 he and his student Annalisa Lattes were successful in this seminal effort. Using an integrated waveguide Mach-Zehnder interferometer that could be imbalanced by the nonlinearity induced by the control pulse in one arm, they demonstrated the first ultrafast all-optical switch. This approach inspired further developments of such architecture over several decades with a variety of material technologies.

    In 1980 I joined Hermann in the MIT Research Laboratory of Electronics, where we grew laser and optics research into a successful synergism of theory and experiment, particularly in the area of ultrashort-pulse and ultrafast-phenomena optics.

    Also in the 1980s, Hermann returned to his interest in noise. During a second sabbatical at Bell Labs, he teamed up with Jim Gordon (NAE 1985, NAS 1988) to quantify the effects of amplifier noise on the propagation of soliton pulses in optical fibers. They found that the frequency jitter caused by the amplified spontaneous emission noise of the amplifiers produced, in the presence of group velocity dispersion, random shifts in pulse arrival time after transmission through the fiber.8 Accounting for and overcoming this “Gordon-Haus effect” has remained an important aspect of the development of ultra-short-pulse optical fiber devices for timing and synchronization. Hermann and postdoc Antonio Mecozzi subsequently showed that simple filtering can limit the random frequency walks so that the timing jitter only grows linearly rather than with the cube of the distance.Hermann returned to this topic a few years later with a comprehensive review of both the theoretical aspects and technological challenges of solitons in communication systems.

    In 1986, in a definitive and oft-cited paper, Hermann and Yoshihisa Yamamoto addressed the preparation, measurement, and information content of quantum optical states in general.10 Hermann and Yinchieh Lai extended the analysis of quantum noise, specifically and in detail, to solitons in optical fibers.11 One motivation expressed in these works was the possible manipulation of noise by “squeezing” it out of one variable of the optical field and into another — complementary — variable. Hermann and his student Keren Bergman ultimately achieved this objective in a fiber interferometer experiment in which the amplitude noise of signal pulses was squeezed below the classical shot-noise limit by more than 3 dB.12

    With the emergence in the 1990s of fiber lasers as potentially important sources of femtosecond pulses, Hermann again led the way in both theory and experiment. He developed the theory of additive-pulse mode locking to guide the use of the self-phase modulation nonlinearity in fibers for the creation of artificial instantaneous saturable absorbers (pulse shapers). He and his student Kohichi Tamura used this approach to achieve femtosecond pulses with a soliton laser13 and then extended it to the demonstration of a new configuration, the stretched-pulse laser.14 This latter system, which achieved shorter pulses and higher powers, was soon licensed to industry and remains the model for many commercial systems.

    In 2000, at the age of 75, Hermann brought together all his insights and theory formulations on ultra-short-pulse generation in a comprehensive review article titled simply “Mode-Locking of Lasers.”15 Meanwhile, his theory of stretched-pulse mode locking was providing the foundation for modeling femtosecond solid-state lasers, which subsequently achieved pulse durations of only a few cycles of light and created the ability to explore the next ultrafast frontier — the attosecond time domain.

    In the late 1990s, as interest developed in optical nanostructures and photonic crystal behavior, Hermann’s profound understanding of guided-wave phenomena and coupled-mode theory put him at the center of these emerging technology efforts as well. He was a strong believer that the most practical approach to densely integrated photonic circuits was through the development of high index-contrast waveguide devices and circuits. He took on new students, obtained funding, and with his team began inventing and demonstrating an array of new integrated silicon photonic devices for filtering, interconnections, polarization rotation, and input/output coupling. He and his student Christina Manolatou turned her Ph.D. thesis into a book, Passive Components for Dense Optical Integration (Springer, 2002).

    Hermann was known at MIT as an extraordinary teacher for both his dynamic in-class personality and his course-defining textbooks. With James Melcher (NAE 1982) he wrote Electromagnetic Fields and Energy (Prentice-Hall, 1989) for the MIT electrical engineering undergraduate course in electromagnetics for which, in successive semesters, he alternately lectured and taught sections. In teaching classes he would gesture dramatically, convey awe of the subject and the phenomena it explained, and lecture without notes. In living by the lesson he had learned on the train from Slovenia to Austria as a refugee, it helped that he had the ability to rederive everything from the ground up, on the spot, if a problem arose.

    For his yearly graduate course, Hermann had written the textbook Waves and Fields in Optoelectronics (Prentice Hall, 1984); and in 2000, at the age of 75, he created yet another course and corresponding major text, Electromagnetic Noise and Quantum Optical Measurement (Springer, 2000), which unified his life’s work on the theory and experimental manifestations of noise.

    Hermann’s substantial contributions were recognized throughout his career, starting with a Guggenheim Fellowship for Natural Sciences in 1959. He subsequently received the IEEE James H. Mulligan Jr. Education Medal (1991) and, from the Optical Society of America, the Frederic Ives Medal (1994) “for his fundamental and seminal contribution to the understanding of quantum noise in optical systems and for a lifetime of dedication to science and engineering education.” In 1995 President Bill Clinton presented him with the National Medal of Science “for his fundamental and seminal research contributions to the field of quantum electronics, noise and ultra-fast optics; and for his service to the engineering profession through teaching.”

    Hermann was always about much more than science and engineering. He inspired colleagues with the stories of his experiences as a young man and with his command of literature, history, and art. He led students to lectures on diverse subjects, to local museums and art exhibits, and on personal tours of campus sculpture. When away from MIT at conferences, he regularly organized long hikes, mountain climbs, and excursions for lake swimming.

    He cherished the successes of his students and friends and celebrated them with a champagne toast or a dinner out. The parties that he and his wonderful wife Eleanor (Lennie, née Laggis) hosted at their home have left generations of students and visitors with fond memories. At their annual summer barbecue, Hermann, with a towel rolled under his arm, would lead the charge to the neighborhood pool for the obligatory predinner swim before returning to cook chicken over a wood fire. Each year there was also a Christmas party, often with one of the students playing the piano for singing and always with Hermann’s famous punch bowl.

    He and Lennie  vacationed with their children on Nantucket and later in Hawaii, where their two sons settled. The couple also kept a family apartment in Vienna, Austria, where the Haus name was still recognized. Hermann’s grandfather, Anton von Haus, had been the commander in chief of the Austro-Hungarian naval fleet under Emperor Franz Joseph during World War I.

    Well into his 70s, Hermann showed no signs of slowing down. After officially retiring at the age of 70 he continued to teach, wrote a new 562-page textbook, Electromagnetic Noise and Quantum Optical Measurements (Springer-Verlag, 2000), and maintained an active and forward-looking research program with graduate students. His sudden death, during his usual bicycle commute home, stunned and saddened all who knew him.

    Not long thereafter, MIT created the Hermann Anton Haus Room for seminars, conferences, and group meetings at the Research Laboratory of Electronics. And the Research Laboratory of Electronics established the annual Hermann Anton Haus Lecture to honor his memory and bring distinguished speakers to the laboratory so as to continue the openly collaborative dialogue that Hermann promoted throughout his life.

    Hermann and Lennie, who outlived him for five years, were married for almost 50 years. They had four children — Bill, Stephen, Cristina, and Mary — and were devoted grandparents to four grandchildren and two step-grandchildren.

    Adapted with permission from Biographical Memoirs of the National Academy of Sciences (available online at The NAS memoir includes some additional details and citations. The author also refers readers to an autobiographical manuscript that Hermann prepared shortly before his death for the 2003 Gordon Conference on Nonlinear Optics; the resulting article was published in 2004 by the Journal of Modern Optics 51:1873-88.
    1Haus HA. 1954. Equivalent circuits for a passive nonreciprocal network. Journal of Applied Physics 25(12):1500-02.
    2Haus HA, Adler RB. 1959. Circuit Theory of Linear Noisy Networks. Cambridge: MIT Press.
    3Haus HA, Mullen JA. 1962. Quantum noise in linear amplifiers. Physical Review 128(5):2407-15.
    4Freed C, Haus HA. 1965. Photoelectron statistics produced by a laser operating below the threshold of oscillation. Physical Review Letters 15(25):943-46.
    5Haus HA. 1975. Theory of mode locking with a slow saturable absorber. IEEE Journal of Quantum Electronics 11(9):736-46.
    6Haus HA. 1975. Theory of mode locking with a fast saturable absorber. Journal of Applied Physics 46(7):3049-58.
    7 Haus HA, Shank CV. 1976. Asymmetric taper of distributed feedback lasers. IEEE Journal of Quantum Electronics 12(9):532-39.
    8Gordon JP, Haus HA. 1986. Random walk of coherently amplified solitons in optical fiber transmission. Optics Letters 11(10):665-67.
    9Haus HA, Mecozzi A. 1992. Long-term storage of a bit stream of solitons. Optics Letters 17(21):1500-1503.
    10Yamamoto Y, Haus HA. 1986. Preparation, measurement, and information of optical quantum states. Reviews of Modern Physics 58(4):1001-20.
    11Lai Y, Haus HA. 1989. Quantum theory of solitons in optical fibers, I and II. Physical Review A 40(2):844-866.
    12Bergman K, Doerr CR, Haus HA, Shirasaki M. Sub-shot-noise measurement with fiber-squeezed optical pulses. 1993. Optics Letters 18(8):643-645.
    13Haus HA, Ippen EP, Tamura K. 1994. Additive-pulse modelocking in fiber lasers. IEEE Journal of Quantum Electronics 30(1):200-208.
    14Haus HA, Tamura K, Nelson LE, Ippen EP. 1995. Stretched-pulse additive pulse mode-locking in fiber ring lasers: Theory and experiment. Journal of Quantum Electronics 31(3):591-598.
    15In IEEE Journal of Selected Topics in Quantum Electronics 6(6):1173-85 (2000).