A&O READING – The caddus worm in Leonardo 2004



(archive copy)

The Wonderful Caddis Worm: Sculptural Work in Collaboration with Trichoptera

Hubert Duprat and Christian Besson[i]   (2004)[ii]

LEONARDO. https://www.leonardo.info/isast/articles/duprat/duprat.html

ABSTRACT.  Since the early 1980s, artist Hubert Duprat has been utilizing insects to construct some of his “sculptures.” By removing caddis fly larvae from their natural habitat and providing them with precious materials, he prompts them to manufacture cases that resemble jewelers’ creations. Information theory, as explained by biologists such as Jacques Monod and Henri Atlan, helps us understand what seems to be the insect’s aesthetic behavior. The activities of the caddis worm, as manipulated by Hubert Duprat, are prompted by the “noise”—beads, pearls and 18-karat gold pieces—introduced by the artist into the insect’s environment. This article is based on a conversation between the artist and art critic Christian Besson.


Christian Besson: Your work seems to be fueled by an insatiable curiosity about science. The world you have created around you—your office, the things you collect and so on—seems to be a kind of cabinet de curiosités,such as existed during the Renaissance, a Wunderkammer—a room of wonders—as the German term so nicely puts it.

Hubert Duprat: The collectors who created thoseWunderkammern were driven by a feeling that I myself experience about art. It dates back a long way. I spent my boyhood and teenage years in the countryside, where I hobnobbed with hunters and fishermen. Very early on, I had a keen interest in archaeology and the natural sciences. I made early observations in aquaria, where I installed water scorpions, water-striders, newts, tadpoles, pond skaters, planorbid snails and, right at the outset, caddis worms—Trichoptera (see Appendixes Aand B).

Ch.B.: Of all of your work, what you do with these caddis worms is the strangest. They are small insect larvae living in streams and ponds, much sought after by trout fishermen. The art world knew nothing of them before you.


H.D.: I have a passion for geology. In 1979, I came upon gold panners at the Ariège river in southwestern France. I then had the idea of providing the caddis fly larvae—the worms that I was already rearing—with gold spangles. I watched how the worms incorporated these spangles into their cases; when I took away a worm’s case, I noted that it would make another one right away, this time made entirely of spangles. This experiment led, in 1983, to my patent application at the French Patent Office [1].

Ch.B.: The cases made by the caddis worm (under your guidance) are something between insect artifact, jewel and sculpture. These objects are works of art and, at the same time, products of an intrinsically scientific experiment. They can be seen in two ways, like a Janus bifrons. As art pieces, they are a kind of assisted ready-mades, found objects altered and promoted to the rank of works of art, hybrid formations such as appeared in large numbers in the 1980s. As scientific experiments, they are evidence of an unexpected interdisciplinarity; they also raise real epistemological issues. But before we tackle these, we should describe the experiment in detail, as well as the way it has been carried out.

H.D.: The larvae I use belong to the families Limnephilidae, Leptoceridae, Sericostomatidaeand Odontoceridae, with a preference for the Limnephilid genera Potamophylax and Allogamus.I collect the larvae from January to April, in low- and medium-altitude mountain areas, and keep them in an aquarium where the water is oxygenated, circulated and kept at 40% C—this artificial winter prolongs the larvae’s period of case-building and delays the process of nymphosis. I remove the cap of the larva’s natural case at the rear end, and push the larva, which usually stays affixed to the case by means of its two hind hooks, gently forward with a blunt-tipped instrument. Pressure applied to the last ring of the case causes the larva to release its grip. Essentially, this in vitro experiment involves the modification of the larva’s natural habitat and, more precisely, the replacement of the building materials ordinarily found by the larva (sand, small bits of gravel, sprigs of plants, the shells of planorbid and other water snails) with new materials. To begin with, I put the insect in a gold-filled environment for as long as it takes the creature to form a rough case. The larva must be able to move around in its new case and be picked up without any risk of breaking the fragile construction. First, I only provided the larvae with gold spangles, but then I gradually added beads of turquoise, opal, lapis lazuli and coral, as well as rubies, sapphires, diamonds, hemispherical and Baroque pearls, and tiny rods of 18-karat gold.



Fig 1

The larva connects the materials with silk thread from inside the tube, using a spiral movement, and then upholsters the inside of the case with a lining also made of silk. Because I keep each type of material in a separate tank, it is possible, by monitoring the time a larva stays in a given tank, to get it to make a ring of a specific material. Also, by damaging the insect’s case (including the lining) in a specific area, it is possible to prompt the creature to repair its case by placing a specific material precisely where the damage was inflicted.


Fig 2

Ch.B.: The building activity of the insect is thus not only sidetracked, but also painstakingly channeled to your aesthetic purpose.

H.D.: The creatures may not use all the materials with the same facility, some may prefer pearls or beads, for example. A tube that was manufactured in one season may be lived in by a new larva the following year. This larva will extend the tube or even produce a new case, which it will eventually separate from the original one.


Fig 3

Ch.B.: Christian Denis [2] notes that because the dimensions of a borrowed case are rarely ideal, the larva uses it as a basis for the construction of another one of a more suitable size.

H.D.: Other creatures than those listed above are certainly also suitable for this experiment. One season I was able to get larvae of Phryganeidae to build mineral cases, even though, in the natural state, they make their cases exclusively of plant matter.

Ch.B.: This raises the question of whether the fact that this experiment is carried out in an exhibition alters its results.

H.D.: Dragging its industriously made case, the insect is shown at eye level in the exhibition, in a small aquarium, duly supplied (behind a screen) with running water, which is cooled and oxygenated. At night, the light-shy larvae are again placed in the various tanks containing the materials necessary for further construction of their cases. What is thus exhibited, artistically speaking, is the (temporary) result [3] of the experiment—not the actual experiment itself. In this respect, I have not turned the caddis worms into circus animals, nor have I put the manufacturing process on display. (To show the process, notes Denis [4], one would have to film the larvae, because they do not build on command).


Fig 4


Ch.B.: The idea of this type of in vitro experiment has cropped up among several professional entomologists, and even among “nature lovers.”

H.D.: This subject fascinates me, so I have checked every possible source—needless to say, I was clueless about the topic when I embarked on this project. The oldest reference I managed to find to an experiment with the caddis worm is in the writings of François-Jules Pictet, an entomologist from Geneva. Pictet carried out “a few tests to see howfar it is possible to get species to work with materials which are not theirs” [5]. He recorded in 1834 that he gave mineral materials to larvae used to working with plant materials and vice versa. The extremely lyrical Jean-Henri Fabre [6], one of the founding fathers of ethology, describes his observations on and experiments with the species Limnephilus flavicornis. He gave the insect grains of rice (at that time a material that was not commonly used) and was duly thrilled by the resulting “consistency of the work.”

The first reference to an aesthetically unusual result occurred in 1923, with Canon C.-H. de Labonnefon. He provided the insect with little glass beads “which gave its abode the many-coloured appearance of a magnificent Harlequin’s costume” [7]. I came across this bead story again in the Traité de zoologie, edited by Pierre Grassé in 1951 [8], and in a book published in 1956 by Gerald Durrell [9], who describes an experiment carried out in Greece.

More recently, there have been experiments with more precise methods than Fabre’s [10]. Polish researchers, for example, have focused on the adaptive building behavior of Trichoptera placed in situations where the case was damaged, thus calling for repair. I should mention above all Denis’s outstanding dissertation [11], which includes, in particular, certain observations on the experimental alteration of the natural habitat (gauge of sand grains, brightness of light, etc.). This researcher was also the first to work from the period of egg-laying, whereas all other experimenters worked with older larvae to rebuild existing cases. But the aesthetic nature of the outcome is of no concern in all recent scientific literature.

The Metaphor of the Craftsperson

Ch.B.: It is a fact that this aesthetic side belongs to the dawn of entomology. By its very stature, the insect craftsman dominates all of the entomological literature of the nineteenth century—scientific and literary alike—from Réaumur to Maeterlinck. From the divine architect in Plato’s Timaeus,by way of Aristotle’s industrious nature, we steal towards the insect builder, ingenious constructor of its own abode, without abandoning the metaphor of the craftsman. The image of the craftsman is a persistent one. It permeates all of Fabre’s work. When he describes the manufacture of the case by the caddis worm, he is unable to refrain from comparing the worm to a bricklayer:

When the mason uses bricks to construct the narrow flue of a factory chimney, he stands in the middle of his turret, and gradually lays new courses by revolving on his own axis. The caddis worm does the very same thing [12].

The rotation of the worker worm here replaces the spinning clay on the potter’s wheel. According to Étienne Souriau [13], among the aesthetic factors occurring in nature, the insect and the plant share style, combinative inventiveness and the ability to feign, but the insect has the specific ability to produce artifacts. In his delightful book, Le Sens artistique des animaux, Souriau devotes an entire chapter to the insect craftsman and an examination of the potter wasp. He finds the wasp “sensitive to the right shape” and refers to Plato’s Timaeus as the “metaphysics of the potter’s art.” As a faithful Platonist, he supposes that “in the hereditary programme of our insect’s action, a model is implied, in some way or other.” Metaphysics triumphs!

The fact is that it is easy to project some design, and we implicitly postulate some intent, in such types of animal behavior as that of the caddis worm. Even if we contrast Aristotle with Plato, it is hard to dodge this a priori. Robert Lenoble noted that, when we observe the stupid insect, we cannot stop ourselves from subscribing to the contemplative artificialism of Aristotle:

Nature manufactures stones, animals and plants the way the potter turns his vase. The theory of the four causes is the rationalization of this image: like the craftsman, Nature takes a variety of matter (material cause) and imposes a form upon it (formal cause) with the help of an instrument (efficient cause); the whole operation being undertaken and carried out with an eye on the result (the final cause is the most important and the only explanatory cause in the last resort) [14].

Abandoning Plato, the image of the potter and the postulate whereby the potter copies a model, decrying the over-humanized descriptions given by Fabre and the whole of literary entomology—in a nutshell abandoning idealism—it is thus hard to sidestep final causes.

Even Kant does not altogether allow us to restrict these metaphysical digressions.

Art is distinct from nature just as “making” (facere) is from “doing” or “causing” in general (agere) and [just as] the product of the consequence of art is distinct as work (opus) from the product of nature as effect (effectus[15].

Kant’s distinction between works of art and those of nature leaves us in a quandary. The production of the artifact within nature herself poses a problem—even more so when an aesthetic aspect is involved. Whether the insect is a craftsperson or whether, more generally, nature is a creator of forms, the consideration, within nature, of an aesthetic dimension is the stumbling block of science.

One of the answers of contemporary biology to this problem involves the notion of teleonomy put forward by Jacques Monod:

All artefacts are the product of a living being, which thus, and in a particularly evident way, expresses one of the basic properties which characterize all living beings without exception—the property of being objects endowed with a projectwhich they at once represent in their structures and accomplish by their performance (such as, for example, the creation of artifacts) [16].

Monod,moreover, acknowledges that the teleonomic character of living beings is in contradiction with the objectivity of nature, and that this epistemological contradiction is the core problem of biology. Kant limited finalism by positing the concept of the finality of nature as a regulatory, and altogether subjective, principle, helping to steer research. For current biology, the concept of teleonomy would replace both erstwhile finalism and Kantian teleological judgment. According to Henri Atlan,

Events and the future forms towards which the organism seems to head are in fact contained at the outset, in a coded way, in the nucleotide sequences of the DNAs of the genome [17].

Contemporary biology explains for us, in this way, the illusion we have when we observe the insect pursuing its end purpose as it builds its case.


Ch.B.: It seems to me that there is a second issue involving the innovative ability that the caddis worm seems to have when it becomes a goldsmith and starts to manufacture jewelry. Not only does it produce a case based on a hereditary model, but it also demonstrates creative behavior. Denis has the opposite view: for him, the caddis worm’s acceptance of new materials depends on their resemblance in size and shape to those normally used. So there is no innovation on the part of the insects, for whom new materials are, when all is said and done, ersatz substitutes, which only our eye regards as aesthetic. This is worth discussing.

H.D.: An American entomologist, Charles T. Brues [18], describes the curious observation he made in a river in northern Nevada. Among all the little particles of sand and minerals swept along by the water, the Trichoptera make meaningful selections of bright blue opals—in other words, the most conspicuous or garish materials. Brues considers this phenomenon to be noteworthy, since the insect’s vision seems to play a part in the process, whereas it was ordinarily thought that materials were evaluated only on the basis of their weight and what they felt like.

Ch.B.: This observation runs counter to current interpretations of mimetism that involve a decoy or lure function. Here, as in your aquaria, you cannot say the caddis worm is trying to hide itself in its case—quite the contrary. Unusual conditions determine new—if not to say, aberrant—behavior. Your artistic involvement likewise boils down to plunging the caddis worm into unusual conditions and, in so doing, triggering an unforeseen reaction.

Can we attribute to the insect manipulated by you, the artist, an aesthetic pattern of behavior? If we refer to information theory, as biologists have in recent years [19], perhaps we can find a way out of the impasse. Atlan, who has worked in this area, uses the notion of noise, borrowed from the theory put forward by Shannon and Weaver [20], to describe the innovation:

Absolute novelty stems from the indeterminate character of stimuli which thus play the part of random upheavals in the system which they affect. The acquisition of new knowledge by experiment is a specific case of information growth under the effect of noise [21].

Atlan links the illusion of final cause to the structure of our mind (which is very Kantian):

Situations do exist where we experiment with a sort of local inversion of time. When we become involved in a voluntary act, a series of gestures is the result of our conscious will and is directed towards the goal that we wish to attain, in such a manner that, in a certain way, the series of events seems determined by the final causes [22].

But another type of situation exists where entropy decreases and complexity increases:

We are dealing with a local inversion of time in so far as there is a local reduction of entropy. [It] plays a part in the—unconscious—physico-chemical processes of self-organization when the principle of “complexity by noise” is at work [23].

Your activity as an artist, upsetting the ordinary ethology of the insect, seems to me to be the same thing as introducing a noise, complicating itsUmwelt and producing a response. In your diversion of the caddis worm’s behavior, in your artistic manipulation, the effect is twofold. From a biological viewpoint, a random event triggers self-organization. From a human viewpoint, the experimenter’s intent produces this effect. These two types of final cause-effects (time inversion) are combined in the in vitro experiment.

Is the caddis worm’s precious case the work of the insect or the work of the artist? This is not the right question. The contradiction can be resolved by the differing viewpoints. According to the first view, the caddis worm owes nothing to the artist (who is simply the author of one noise among the thousands of other noises in its environment). According to the second view, the caddis worm is merely the executor of the artist’s project. The artistic statement plays on the confusion of the two levels by overlaying the two perspectives. The aesthetic result (at once natural and artistic) turns the caddis worm’s case—which is more than an assisted ready-made or a “diversion”—into a doubly exposed object, like a double exposure: a scientific-cum-artistic palimpsest.

H.D.: So we are still a long way from bringing the subject full circle. I have also delved into the literature on mimetism. In it, the caddis worm is invariably mentioned.

Ch.B.: We will discuss that in another article—where, incidentally, we might talk about your other work [24]. We could illustrate how close your art is to the aesthetic thinking of Roger Caillois, who was very interested in mimetism. He was a writer who made no separation between art and sciences. And, like you, or so it seems to me, he was looking for what he called “adventurous coherences.”


Appendix A: The Trichoptera

By Hubert Duprat and Christian Besson 

The order Trichoptera (once treated as a suborder of the Neuroptera) is included in the superorder Mecopteroidea. It numbers 895 species in Europe (based on the works of Botosaneanu and Malicky, Limnofauna Europaea, 1978 
[25]). According to these authors, the term caddis fly applies to all of the Trichoptera but is also used for just the one family making up the type of the order. The insect’s scientific name is derived from the imago—the adult insect, which tends to be active at dusk—the wings of which are usually covered with tegumental hairs.

Together with campodeiform (bristletail) families, the larvae of which are free-moving and make trap-nets, there are 12 families (in France) with distinctive eruciform (caterpillar-shaped) aquatic larvae that manufacture a case made of mineral or plant matter or a combination of both. The larvae dwell in the case, but do not fasten themselves to it. The order is holometabolic, which means that the insect’s whole body undergoes metamorphosis during nymphosis. Nymphosis takes place inside the case, which is closed for this purpose. The nymph, like the larva, has tracheal gills. The buccal apparatus is of the grinding type, and the labium bears an outlet for two silk-producing glands. The tube-shaped case has several functions. Minerals act as ballast for the insect in flowing water. Plant matter enables the insects to trap an air bubble and develop on the surface of stagnant water. Silk protects the gills and controls the amount of water from which oxygen is extracted 
[26]. Finally, because of its mimetic appearance, the case protects the larva from predators.

Appendix B: The Historical Development of the Order Trichoptera

By Irina Soukatcheva

The Trichoptera are insects that resemble butterflies [more precisely, according to Christian Denis, primitive butterflies that do not look like the more recent, showier butterflies], but unlike the latter, they have an aquatic larva. The larvae of Trichoptera often live in skillfully constructed “mobile homes.” This distinctive feature appeared in Trichoptera after lengthy historical evolution. The oldest representatives of this biological order were widespread on Earth some 200 million years ago. In those remote times, early specimens of Trichoptera only vaguely resembled their modern descendants. As for the larva, it did not build shelter (cf. my book, Le développement historique de l’ordre des Trichoptères, Moscow, 1982 [in Russian]). 

As the larvae evolved, they first and foremost learned to construct fixed shelters (galleries) and only developed transportable tubes at a later date. These appeared some 150 million years ago, when the larvae were no longer carnivorous and turned into vegetarians. This turning point meant that the great mobility of the larvae was no longer of use, but it also called for an effective defense against predators. This defense could be provided by stout and moveable cases, built with benthonic particles (on the beds of water courses), mineral and organic alike.

Among cases unearthed, the oldest have been discovered in large numbers in Central Asia. They belong to the late Jurassic and early Cretaceous period. Their construction is somewhat primitive—a not very meticulous assembly of disparate grains of sand of many shapes, fragments of plants, shells, bits of fish bones and so on. Subsequently, as their building instinct became more complex and developed, the larvae learned how to build increasingly diversified and well-made shelters. The most accomplished cases date from the early Cretaceous. They appeared, in other words, some 50 million years after the test models. At that time, case-building Trichoptera were already widespread throughout the world, and their diversity was comparable to their present state. 

Then, as now, the larvae used not only grains of sand, mica, and mollusc and crustacean shells, but also water weed, small ligneous fruit, needles and plant particles, eaten away or otherwise. They are also very widespread in more recent deposits, where the oddest of specimens occasionally come to light. Thus, in 1829, in Oligocene deposits in the Auvergne, Professor P.M. Serres discovered magnificent constructions made with cleverly selected gastropod shells. They are so numerous that they form whole layers of indusial limestone. Because their apprenticeship has spanned tens of millions of years, the larvae have learned not only how to handle different materials, but also how to arrange these materials in very canny ways, following a precise method (starting with tight parallel rows and developing to more sophisticated systems). In addition, so that the cases can float (or not), the larvae sometimes cover them with larger than usual grains of sand, as well as with mollusc bodies and even long twigs and blades of grass. 

Hubert Duprat’s striking and truly fantastic experiments clearly illustrate the outstanding level that has been reached by the caddis fly larva’s building art. Behind these distinctive and accomplished creations of nature lies a whole history of the evolution of the building instinct, dating right back to the age of the dinosaurs. It is a sad fact, today, that an environmental threat hangs over the caddis flies, whose larvae are very demanding in terms of water quality and cannot survive in polluted waters. So there is a risk that future generations will not have a chance to see these miracles of nature, which are so fragile 

References and Notes

1. French Patent No. 83 02024. 

2. Christian Denis, Contribution à l’étude du comportement constructeur des larves de Trichoptères et problèmes relatifs à l’édification du fourreau (Paris: Crepin-Leblond et Cie, 1968) (dissertation presented at the Faculty of Science at Rennes).

3. The result is temporary because the case degrades over time. 

4. Letter to Hubert Duprat, 17 November 1995. 

5. François-Jules Pictet, Recherches pour servir à l’histoire et à l’anatomie des phryganides (Geneva: Abraham Cherbuliez, 1834) pp. 117–119.

6. Jean-Henri Fabre, Souvenirs entomologiques,7th series, XX (Paris: 1879–1907) p. 282.

7. Canon C.-H. de Labonnefon, Croquis entomologiques (Paris: Maison de la bonne presse, 1923) p. 56. 

8. Raymond Despax, “Order des Trichoptères,” in Pierre Grass é, Trait é de zoologie, Vol. 10, Fasc. 1, (Paris: 1951). 

9. Gerald Durrell, My Family and Other Animals(London: Hart-Davis, 1956).

10. For a complete bibliography, see Denis [2].

11. Denis [2].

12. Plato, Timaeus, 50 c–e.

13. Etienne Souriau, Le sens artistique des animaux (Paris: Hachette, 1965). 

14. Robert Lenoble, Esquisse d’une histoire de l’idée de nature (Paris: Albin Michel, 1969) p. 82 ff.

15. Emmanuel Kant, Kritik der Urteilskraft(1790). Quoted from the French translation by Philonenko (Paris: Vrin, 1965) p. 134. 

16. Jacques Monod, Le hasard et la nécessité(Paris: Le Seuil, 1970) p. 22. 

17. Henri Atlan, Entre le cristal et la fumée (Paris: Le Seuil, 1979) part 1, p. 1. 

18. Charles T. Brues, “Jewelled Caddis-Worm Cases,” Psyche 37, No. 4, 392–394 (1930). 

19. Henri Atlan, L’organisation biologique et la th éorie de l’information (Paris: Hermann, 1972).

20. C.E. Shannon and W. Weaver, The Mathematical Theory of Communication (Urbana, IL: 1949). For an application of this theory to aesthetics and for an explanation of the assimilation of the concept of information to that of originality, see Abraham Moles, Théorie de l’information et perception esthétique (Paris: Flammarion, 1958), in particular chaps. 1, 5 and 6.

21. Atlan [17] p. 170.

22. Atlan [17] p. 158.

23. Atlan [19] p. 172.

24. Christian Besson, “La Phrygane, la merveille et le monument,” in Hubert Duprat, exh. cat. (Paris: Hôtel des arts, 1991) (includes English translation). Duprat’s work was recently exhibited in a one-man exhibition at the Picasso Museum, Antibes, 9 April–15 June 1998. The exhibition was accompanied by a catalog with contributions by R. Recht, M. Fréchuret and S. Bann.

25. Botosaneanu and Malicky, Limnofauna Europaea (Amsterdam: Illiés, 1978). 

26. Some cases are made only of plant matter; others only of minerals. Some cases combine both plant matter and minerals. All cases are bonded by silk.

27.Irina Soukatcheva is a researcher at the Institute of Palaeontology at the Russian Academy of Science. This text, translated, revised and corrected by Kiril Tchekalov, appeared in Nous sommes les pensées d’un ange, exh. cat. (Moscow: Avant-garde, 1995).

Manuscript received 12 February 1997.

Figure Captions

1,2,3,4: Aquatic caddis fly larvae with cases, gold, pearls, precious stones, 2–3 cm each, 1980–1996. (Photos: H. Del. Olmo)


[i] Hubert Duprat (artist), rue du Four, 34270 Claret, France.  Christian Besson (philosopher, art critic), 12, rue de Mazy, 21160 Marsanny-la-Côte,France.  Translated by Simon Pleasance

[ii] Updated 17 November 2004.  http://leonardo.info