Until recently, the Leiden ethology department was housed in a tall building seven stories high. On top was the tropical aquarium. When - a long time ago, say 1964 - on Sunday afternoons I went all the way up in the elevator to feed the fishes, a thin, spectacled face with a neat chin-tuft might show and stare at me through the narrow window of the door at the fifth. Keith Nelson preferred the weekends for his experiments with sticklebacks, because then there were less noises and vibrations that might disturb his experimental fish. This arrangement had the additional advantage that he was free in the middle of the week when everybody was at work. He loved to take his car, accompanied by Nancy and their baby daughter Johanna, to explore and criticise The Netherlands.
His driving style was sort of restless: steer, brake, accelerate as an uninterrupted series of actions. Simultaneously, within the privacy of the car, he gave vent to all his little irritations: "you ugly milkboor, get out of my way" or "dirty old man, I'll flatten you!"
His mind worked similarly, springy, keen and sharp, seemingly chaotic but at the same time cutting a straight path for himself through the jungle of other people's objections or current theory. One couldn't fail to appreciate a touch of genius in him. His experiments with the stickleback's 'creeping-through' behaviour caused a minor revolution in the study of this animal's behaviour (1). So far, 'creeping-through-the-nest' - an action that the male stickleback performs from time to time in order to keep the tunnel through his nest open - had been considered as a product of the spontaneous fluctuations of aggressive and sexual motivations. The 'creeping-through' act marks, in that view, the moment when the sexual motivation overcomes the aggressive. Immediately following the act, the male is super eager to court a female; so much so that he will respond to things that have only a remote resemblance to a female stickleback such as a snail or air bubbles rising to the surface.
Individual 'creeping-throughs' are separated in time in the order of hours. As to what determines the length of the interval our specialists (2) knew that the next 'ct' is delayed in the presence of a conspecific male (enhancing aggression), or expedited by a female. Fair enough. Keith boldly reversed the argument and hypothesised that 'ct' had its own, autonomous cycle with an internal clock that reigned the dynamics of the sexual and aggressive motivations instead of being its result.
In order to test his hypothesis, he designed a simple model with two factors: an Excitation (E) and a Threshold (T), both of which decrease with mathematical precision as long as they are not stimulated. Where the curves of the two processes cross, the fish creeps-through and T is elevated to a fixed value. With this model he could predict the moment of the next 'ct' one and a half to two hours ahead with a deviation of a mere 1 or 2 minutes; an unheard-of precision for a behavioural process! Conducting the experiment was not difficult, but it took a lot of patience. One had to measure the duration of two consecutive 'ct' intervals. The model then predicted the length of the third. You may now understand why Keith worked in the quiet hours: a minimum of stimulation that might influence the course of his E's and T's. His staring through the elevator window wasn't just curiosity but an expression of irritation that somebody dared to come along. To see your experiment being spoilt when you have already invested some hours in it can be very frustrating! Inclined to "flatten" anybody who did!
Keith's model perfectly fitted the behaviour of the stickleback, and not only that. Similar models describe the circadian sleep cycles of animals and humans and sand bathing in chickens (3). Keith doesn't always get the credit he deserves.
Apart from sticklebacks, he observed some other animals, for instance the Song Thrushes in the Leidse Hout. Early morning he would be ready on the spot with his recording gear to catch the singing of the birds. It resulted in a splendid paper in the Festschrift for Piet Sevenster in 1990 (see note nr 6).
Prior to his coming to Leiden, Keith had received his PhD from the University of California at Berkeley. His thesis comprised an analysis of the behaviour of the Glandulocaudinae (4) (tail with glands), a subdivision with relatively few species of the Characidae, a freshwater fish family with more than a thousand, perhaps thousands of species. "Real fishes", as Keith characterised them, along with other large families such as Cyprinids or Poecilids, and in contrast to such curio's as Sea-horses, Sticklebacks or Mud-skippers.
Modesty was not among Keith's virtues. Sometimes he wore a 'deerstalker' à la Sherlock Holmes, thus implying that others were like Watson, Lestrade or even Athelney Jones. Great sense of humour, though. Laughed a lot about all those diminutives with which Dutch language abounds and when he first had heard the word 'eventjes' he nearly died of it. Together with Ilan Golani and Wolfgang Schleidt he played many a prank at the IEC (5) of Stockholm in 1965 - but that is another story.
He coined the phrase "Dutch Drive Mysticism" to jeer at the Leiden ethologists, because they valued and thought in terms of Lorenz's and Tinbergen's theories of motivation. To be sure, they weren't exceptional; pick any handbook on ethology of the time and you'll find at least a chapter on motivation and 'drive' as well as one on conflict behaviour (when two motivations interact). Keith's PhD thesis broke new soil by introducing stochastic processes in behaviour, that is when behaviour elements follow one another in a perfectly random order. The chance that an element occurs at a given moment is then dependent only on its general probability of occurrence, and not on preceding elements. Such a state of affairs does not tally with the classical notion of underlying motivations that fluctuate slowly relative to the visible succession of behaviour elements; and so Keith opposed the classical view.
In particular, his criticism was aimed at a common error of thought. Suppose one wants to test the hypothesis that aggressive (or sexual, escape or parental) motivation plays a role in behaviour acts A, B or C. Then "what is the best parameter of aggression (mutatis mutandis)?" was an often-heard question. It hid an erroneous assumption, though, that 'aggression' was a visible and tangible entity, the existence of which had been scientifically established; whereas it was really a theoretical construct to explain a collection of facts. As if the aggressive drive is something one may pick up and measure length, weight or activity of, the equivalent of an upper arm bone, a sweat gland or a stomach. Even in these anatomical examples, however, differences between parameters are of a qualitative nature rather than 'better' or 'best'.
And after Leiden? One might have expected a successful career; that man had the capabilities to alter behavioural science, but it went differently. After some years at the University of Maryland and San Francisco State University, he spent a sabbatical at Tel-Aviv University in 1971. There he interacted with many colleagues, particularly with Ilan Golani; and there he must have done most of the necessary computer work for his great analyses of the song of the European Song Thrush for Sevenster's Festschrift (6) and another pioneering study involving the song of Swainson's Thrush published in 1973 (7). Ilan Golani recently recalled how Keith would spread the prints of all the song types on the floor and stand upon a chair to get a good overview.
He worked at the Bodega Marine Laboratory, Bodega Bay, California from c. 1974 to 1992 and published a fair number of articles on the genetics and evolution of various commercially important organisms; not the epoch-making quality of his behaviour papers, but staunch good quality. Many a young revolutionary ends up as an old master, and Keith seemed to follow the trajectory. However, it looks like the revolutionary blood remained. After retirement in 1993 he devoted his life to the Arts, creating colour etchings. Some pictures of them may be found on facebook:
https://www.facebook.com/KeithNelsonArtist
Notes
(1) Nelson, K. 1965. After-effects of courtship in the male three-spined stickleback. Zeitschr.f.Vergl.Physiol. 50: 569-97.
(2) Iersel, J.J.A. van 1953. An analysis of the parental behaviour of the male Three-spined Stickleback. Behaviour Suppl. 3: 1-159.
Sevenster, P. 1961. A causal analysis of a displacement activity (Fanning in Gasterosteus aculeatus L.) Behaviour Supple. 9: 1-170.
(3) Daan, S., D.G.M. Beersma & A.A. Borbély 1984. Timing of human sleep: recovery process gated by a circadian pacemaker. Am. J. Physiol. R161-R178.
Hogan, J.A. 1997. Energy models of motivation: a reconsideration. Applied Animal Behaviour Science 53: 89-105.
(4) Nelson, K. 1964a Behaviouir and morphology in Glandulocaudine fishes (Ostariophysi, Characidae). Univ. of California Publ. Zoology. 75/2: 59-152.
(5) International Ethological Conference.
(6) Nelson, K. 1990. Hierarchical organization, revisited. Neth. J. Zool. 40(4): 585-616.
(7) Nelson, K. 1973. Does the holistic study of behavior have a future? In: P.P.G. Bateson & P.H. Klopfer (eds.) Perspedtives in Ethology. Plenum Press, New York and London.
 |
| Keith Nelson in Leiden c. 1964. (Photo by Nancy Nelson) |
His driving style was sort of restless: steer, brake, accelerate as an uninterrupted series of actions. Simultaneously, within the privacy of the car, he gave vent to all his little irritations: "you ugly milkboor, get out of my way" or "dirty old man, I'll flatten you!"
His mind worked similarly, springy, keen and sharp, seemingly chaotic but at the same time cutting a straight path for himself through the jungle of other people's objections or current theory. One couldn't fail to appreciate a touch of genius in him. His experiments with the stickleback's 'creeping-through' behaviour caused a minor revolution in the study of this animal's behaviour (1). So far, 'creeping-through-the-nest' - an action that the male stickleback performs from time to time in order to keep the tunnel through his nest open - had been considered as a product of the spontaneous fluctuations of aggressive and sexual motivations. The 'creeping-through' act marks, in that view, the moment when the sexual motivation overcomes the aggressive. Immediately following the act, the male is super eager to court a female; so much so that he will respond to things that have only a remote resemblance to a female stickleback such as a snail or air bubbles rising to the surface.
Individual 'creeping-throughs' are separated in time in the order of hours. As to what determines the length of the interval our specialists (2) knew that the next 'ct' is delayed in the presence of a conspecific male (enhancing aggression), or expedited by a female. Fair enough. Keith boldly reversed the argument and hypothesised that 'ct' had its own, autonomous cycle with an internal clock that reigned the dynamics of the sexual and aggressive motivations instead of being its result.
In order to test his hypothesis, he designed a simple model with two factors: an Excitation (E) and a Threshold (T), both of which decrease with mathematical precision as long as they are not stimulated. Where the curves of the two processes cross, the fish creeps-through and T is elevated to a fixed value. With this model he could predict the moment of the next 'ct' one and a half to two hours ahead with a deviation of a mere 1 or 2 minutes; an unheard-of precision for a behavioural process! Conducting the experiment was not difficult, but it took a lot of patience. One had to measure the duration of two consecutive 'ct' intervals. The model then predicted the length of the third. You may now understand why Keith worked in the quiet hours: a minimum of stimulation that might influence the course of his E's and T's. His staring through the elevator window wasn't just curiosity but an expression of irritation that somebody dared to come along. To see your experiment being spoilt when you have already invested some hours in it can be very frustrating! Inclined to "flatten" anybody who did!
Keith's model perfectly fitted the behaviour of the stickleback, and not only that. Similar models describe the circadian sleep cycles of animals and humans and sand bathing in chickens (3). Keith doesn't always get the credit he deserves.
Apart from sticklebacks, he observed some other animals, for instance the Song Thrushes in the Leidse Hout. Early morning he would be ready on the spot with his recording gear to catch the singing of the birds. It resulted in a splendid paper in the Festschrift for Piet Sevenster in 1990 (see note nr 6).
Prior to his coming to Leiden, Keith had received his PhD from the University of California at Berkeley. His thesis comprised an analysis of the behaviour of the Glandulocaudinae (4) (tail with glands), a subdivision with relatively few species of the Characidae, a freshwater fish family with more than a thousand, perhaps thousands of species. "Real fishes", as Keith characterised them, along with other large families such as Cyprinids or Poecilids, and in contrast to such curio's as Sea-horses, Sticklebacks or Mud-skippers.
Modesty was not among Keith's virtues. Sometimes he wore a 'deerstalker' à la Sherlock Holmes, thus implying that others were like Watson, Lestrade or even Athelney Jones. Great sense of humour, though. Laughed a lot about all those diminutives with which Dutch language abounds and when he first had heard the word 'eventjes' he nearly died of it. Together with Ilan Golani and Wolfgang Schleidt he played many a prank at the IEC (5) of Stockholm in 1965 - but that is another story.
He coined the phrase "Dutch Drive Mysticism" to jeer at the Leiden ethologists, because they valued and thought in terms of Lorenz's and Tinbergen's theories of motivation. To be sure, they weren't exceptional; pick any handbook on ethology of the time and you'll find at least a chapter on motivation and 'drive' as well as one on conflict behaviour (when two motivations interact). Keith's PhD thesis broke new soil by introducing stochastic processes in behaviour, that is when behaviour elements follow one another in a perfectly random order. The chance that an element occurs at a given moment is then dependent only on its general probability of occurrence, and not on preceding elements. Such a state of affairs does not tally with the classical notion of underlying motivations that fluctuate slowly relative to the visible succession of behaviour elements; and so Keith opposed the classical view.
In particular, his criticism was aimed at a common error of thought. Suppose one wants to test the hypothesis that aggressive (or sexual, escape or parental) motivation plays a role in behaviour acts A, B or C. Then "what is the best parameter of aggression (mutatis mutandis)?" was an often-heard question. It hid an erroneous assumption, though, that 'aggression' was a visible and tangible entity, the existence of which had been scientifically established; whereas it was really a theoretical construct to explain a collection of facts. As if the aggressive drive is something one may pick up and measure length, weight or activity of, the equivalent of an upper arm bone, a sweat gland or a stomach. Even in these anatomical examples, however, differences between parameters are of a qualitative nature rather than 'better' or 'best'.
And after Leiden? One might have expected a successful career; that man had the capabilities to alter behavioural science, but it went differently. After some years at the University of Maryland and San Francisco State University, he spent a sabbatical at Tel-Aviv University in 1971. There he interacted with many colleagues, particularly with Ilan Golani; and there he must have done most of the necessary computer work for his great analyses of the song of the European Song Thrush for Sevenster's Festschrift (6) and another pioneering study involving the song of Swainson's Thrush published in 1973 (7). Ilan Golani recently recalled how Keith would spread the prints of all the song types on the floor and stand upon a chair to get a good overview.
He worked at the Bodega Marine Laboratory, Bodega Bay, California from c. 1974 to 1992 and published a fair number of articles on the genetics and evolution of various commercially important organisms; not the epoch-making quality of his behaviour papers, but staunch good quality. Many a young revolutionary ends up as an old master, and Keith seemed to follow the trajectory. However, it looks like the revolutionary blood remained. After retirement in 1993 he devoted his life to the Arts, creating colour etchings. Some pictures of them may be found on facebook:
https://www.facebook.com/KeithNelsonArtist
 |
| The Grandchilds with Keith and Nancy c. 2008 (Photo by Hanna Nelson) |
Notes
(1) Nelson, K. 1965. After-effects of courtship in the male three-spined stickleback. Zeitschr.f.Vergl.Physiol. 50: 569-97.
(2) Iersel, J.J.A. van 1953. An analysis of the parental behaviour of the male Three-spined Stickleback. Behaviour Suppl. 3: 1-159.
Sevenster, P. 1961. A causal analysis of a displacement activity (Fanning in Gasterosteus aculeatus L.) Behaviour Supple. 9: 1-170.
(3) Daan, S., D.G.M. Beersma & A.A. Borbély 1984. Timing of human sleep: recovery process gated by a circadian pacemaker. Am. J. Physiol. R161-R178.
Hogan, J.A. 1997. Energy models of motivation: a reconsideration. Applied Animal Behaviour Science 53: 89-105.
(4) Nelson, K. 1964a Behaviouir and morphology in Glandulocaudine fishes (Ostariophysi, Characidae). Univ. of California Publ. Zoology. 75/2: 59-152.
(5) International Ethological Conference.
(6) Nelson, K. 1990. Hierarchical organization, revisited. Neth. J. Zool. 40(4): 585-616.
(7) Nelson, K. 1973. Does the holistic study of behavior have a future? In: P.P.G. Bateson & P.H. Klopfer (eds.) Perspedtives in Ethology. Plenum Press, New York and London.
