The Thorvaldsens Museum Archives

Busts of Socrates and Apollo 1805-06

Two unknown busts of Socrates and Apollo

During the early years of his career in Rome, Thorvaldsen frequently produced marble busts sculpted after ancient models. He did so for three reasons. First, this gave him the opportunity to perfect his marble sculpting technique on the basis of the best classical paradigms. Second: for a neoclassical sculptor, copying ancient busts was a highly practical method of gaining fluency in much of the formal idiom of classical antiquity. Finally, it was Thorvaldsen’s hope that these marble busts, as reproductions of ancient classics, could be sold more easily to the cultured public.

Thorvaldsens Museum possesses four of these copied busts, namely, Homer, A751, Agrippa, A759 and A760, and Cicero, A761. The largest overall commission of Thorvaldsen’s copies of classical busts, amounting to eight pieces in total, was by the Courlandish art collector Theodor von der Ropp in 1804-1805. For an account of that commission, please see the associated Related Article.

This article concerns two other marble copies of classical busts: those of Socrates and Apollo. These two busts will here be attributed to Thorvaldsen, even though they have not previously been counted as part of his oeuvre; even though they are only seldom mentioned in the written sources; even though their physical appearance has not yet been established; and even though the current location of one of them, the bust of Socrates, remains unknown. This pair of busts thus presents us with a host of (modest) unsolved mysteries about Thorvaldsen’s production—to which likely solutions will be presented here.

Thorvaldsen’s bust of Socrates

Given the extant sources, we have no reason to doubt that Thorvaldsen did indeed produce a marble bust of Socrates i 1805-06. This bust is mentioned in two locations. First, a letter dated 30.8.1805 makes clear that the rough carving of Socrates was initiated in Thorvaldsen’s studio in 1805 by one of his assistants:

... ho messo sotto il Socrate come Lei mi lasciò detto, che lo lavora Carluccio, ed il marmo riesce benissimo [I [have] made a start on Socrates, as you have let me say, by the worker Carluccio, and the marble looks very beautiful] ...

Here Socrates is mentioned together with other works that Thorvaldsen’s assistants were then in the process of carving.

Meanwhile, the bust must have been finished by the following year. It appears on a list, dated 18.6.1806, of works that Thorvaldsen had offered for sale with the following announcement:

Professor Thorvaldsen in Rome, Danish sculptor, producing in marble
Among the completed figures, the following are available:
Socrates, bust, for 60 scudi
Justinian’s – – – – – – – – 80 scudi.

The sources thus indicate that a Socrates bust was complete; but what did it look like? To start with, we may safely assume that this Socrates was a copy of a well-known ancient bust; for nowhere else in Thorvaldsen’s oeuvre do we find the sculptor himself creating “original” portrait busts of classical personages. In general, when Thorvaldsen produced portrait busts of figures from antiquity, he copied them from ancient models. As mentioned above, this occured with special frequency early in his career.

With regard to possible classical models that Thorvaldsen might have used as the basis for his Socrates, meanwhile, there are numerous possibilities. To narrow these down, we may consider an additional source in which a Socrates bust appears—namely, the auction catalogue from Thorvaldsens Museum’s 1849 auction of works by Thorvaldsen and by others that were not incorporated into the Museum’s collection. Here we find, as nos. 88 and 89:

88. Bust of Apollo in Marble, Copy of a Classical Bust. On a Round Pedestal. 2 ft. in height.
89. Bust of Socrates in Marble, Copy of a Classical Bust. On a Round Pedestal. 1 ft., 7 in. in height.

In order to solve the riddle of which bust of Socrates Thorvaldsen copied in 1805-06, then, it is crucial to establish whether that copy is the same as the Socrates that was auctioned off in 1849. It will here be claimed that this was the case, for the following reason: it is by no means unlikely that the 1805-06 Socrates bust stood unsold in Thorvaldsen’s workshop for the remainder of the sculptor’s life, and so found its way into the portion of his estate that was auctioned off in 1849. For with the exception of Ropp’s commission, Thorvaldsen was unable to sell the marble copies of ancient busts that he had set on display early in his career at Rome. Those presently in the Museum’s possession were sent from Rome to Nicolai Abildgaard in 1800-02 at the Royal Academy of Fine Arts—all in an unsuccessful attempt to have them sold. What is more: shortly before his death, Thorvaldsen presented his friend Christine Stampe with a Melpomene bust, which was also a marble copy of a classical bust produced around 1800. Melpomene, then, had been standing in his workshop ever since then. The same fate can certainly have befallen the Socrates bust from 1805-06.

Strictly speaking, it remains possible that the Socrates bust that Thorvaldsen is known to have produced in 1805-06 is not identical to the Socrates that was sold in 1849. But in all likelihood, they were the same; on this see also the description below of the parallel fates of Socrates and Apollo.

The model for Thorvaldsen’s Socrates

If we now assume that the Socrates bust auctioned off by the Museum in 1849 was the same as that sculpted in 1805-06, then it becomes possible to identify the original classical model by tracing the bust’s fate following the auction. Namely: Socrates was purchased in 1849 by the railway director Christian M. Poulsen, whose Flensburg mansion, Villa Sollie, was designed and built in 1855-56 by Gottlieb Bindesbøll. From Poulsen, Socrates passed to his brother’s grandson, the bookseller and publisher Paul Hagerup (1889-1971). In 1965, Hagerup presented his bust to Thorvaldsens Museum; but at the time, the Museum was not aware of the written documentation of the bust cited above, and so could not attribute the bust to Thorvaldsen. Nevertheless, the Museum did not rule out the possibility that the bust had originated in Thorvaldsen’s workshop.
At the same time, the Museum did establish that the model for Hagerup’s bust was the Socrates herm that was already present in the Villa Albani in Rome during Thorvaldsen’s day. That bust had been found in 1735, and is characterized by its similarities to the so-called Silenus, a frequently inebriated satyr in the wine-god Bacchus’s entourage.

Sokrates, plaster, 51 cm, The Royal Cast Collection.
Plaster cast after a Roman copy of a Hellenistic original from the 3rd or 2nd century BC, Villa Albani, Rome

Plaster casts of the Socrates bust in the Villa Albani can be found in The Royal Cast Collection of the National Gallery of Denmark, as well as in the Greek cast collection in the Archaeological Collections of the University of Graz (Austria). Following Paul Hagerup’s death, Thorvaldsen’s bust was sold by (Arne) Bruun Rasmussen Auctioneers of Fine Art in December 1971, op. cit. The auction catalogue echoed the Museum’s 1965 assessment of the bust’s origins: “Marble bust, Socrates … The bust has been attributed to Thorvaldsen’s workshop.”
It has not been possible to trace the bust’s current whereabouts.

Thorvaldsen’s bust of Apollo

As indicated above, the Socrates bust was sold together with an Apollo bust at the 1849 auction at Thorvaldsens Museum. At the time, this Apollo was acquired by the Spanish envoy Leopoldo Augusto de Cueto (1815-1901). The bust has remained in Spanish possession ever since; in 2000, it was acquired by the museum of the academy of arts in Madrid, Museo de la Real Academia de Bellas Artes de San Fernando.
It is now evident that this bust is a marble copy of the famous head of Apollo from the equally renowned Giustiniani collection in Rome, which was dissolved over the course of the 19th century. The Giustiniani head of Apollo is now found in the British Museum.

Apollo Giustiniani, British Museum. Photo: © Marie-Lan Nguyen / Wikimedia Commons
Apollo Giustiniani
120-140 A.D., marble, 45 cm
The British Museum, London
Bertel Thorvaldsen: Apollo Giustiniani, Madrid
Thorvaldsen: Apollo Giustiniani,
copy of a classical model, ca. 1805-06, marble, c. 62 cm,
Museo de la Real Academia de Bellas Artes
de San Fernando, Madrid

This leaves the question of whether the history of the bust in Madrid can be traced via sources prior to the 1849 auction catalogue. In fact, this does appear to be the case. On the 1806 list of works in marble that Thorvaldsen offered for sale, we find, together with Socrates:

Socrates, bust, for 60 scudi
Justinian’s – – – – – – – – 80 scudi.

In other words, there was another marble “bust” besides Socrates, in a somewhat larger size (hence the higher price), called Justinian’s.
It is thus reasonable to assume that this Justinian’s bust is identical to Thorvaldsen’s copy of the aforementioned Apollo Giustiniani, now in the British Museum. After all, it was not unusual for famous antiques to be named after the collection or location housing them; the best-known examples are the Farnese Hercules and the Apollo Belvedere. In the present case, we are dealing with an Apollo also known as Guistiniani, which on Thorvaldsen’s list has been altered, or Danified, to Justinians [Justinian’s]; the genitive s [English: ’s] seems only to emphasize that this is a work that belonged to Justinian / Giustiniani.
The possibility that the Apollo Giustiniani played a prominent role in Thorvaldsen’s reception of classical antiquity is supported by a letter from his friend, the Danish archaeologist Georg Zoëga. On January 7, 1792, Zoëga sent a report from Rome back to Denmark—to Nicolai Abildgaard at the Danish Academy of Fine Arts—concerning the sculptor and Academy fellow Peter Leonhard Gianelli: “Among other things, I have suggested [that he buy a cast of] the Giustiniani Apollo head that has just been made; and he agrees with me that it must be purchased, as a classical piece that has drawn so much attention, and about whose rivalry with the Apollo head at the Vatican [i.e., the Apollo Belvedere] so much has been spoken and written.”

At the time, then, the bust of Apollo in the Giustiniani collection was famous enough that it certainly could have been abbreviated as “Justinian’s.” Clearly it was famous enough that Zoëga could recommend it as a classical ideal that was worth acquiring. What is more, much as Zoëga had brought this bust to Gianelli’s attention, it is hardly unthinkable that he might have done the same for Thorvaldsen only a few years later—just as Zoëga generally advised Thorvaldsen about which ideals of classical scuplture were worth imitating.

Further evidence of Thorvaldsen’s interest in Apollo Giustiniani is found in the sculptor’s plaster cast collection, which includes a variant of the bust, L105, differing only lightly from original.

All in all, there should not be much doubt at all that the Justinian’s that appears as a work by Thorvaldsen on the 1806 list is in fact a copy of the classical marble head of Apollo Giustiniani.

As with the Socrates bust, the sticking-point in this argument is of course the question of whether the 1806 copy of Apollo Giustiniani is the same bust as the Apollo sold at auction in 1849. One further indication that it was relates to the size of the busts: in 1849, Apollo was described as 2 feet (62 cm) tall, while Socrates is listed as 1 foot and 7 inches (49,7 cm). This corresponds to the price difference of 80 and 60 scudi, respectively, on the 1806 list, inasmuch as the prices of Thorvaldsen’s busts were determined by their size—on which see the Related Article on the prices for Thorvaldsen’s works.
Naturally, the sizes of the busts are not decisive here. The main argument for the identity of the Apollo on the the 1806 list with the one auctioned off in 1849 corresponds to the argument made previously in the case of Socrates. This is that, although Thorvaldsen offered both busts for sale in Denmark in 1806, they remained in the sculptor’s workshop as an unsold pair throughout the rest of his life, and so recurred as a pair in the catalogue of Thorvaldsens Musuem’s 1849 auction.

Were the busts truly made by Thorvaldsen?

All the same, there remains one reason to doubt that Thorvaldsen was responsible for these busts: at the 1849 auction, they were advertised under the category of “Sculptures by Other Artists.” While at first glance, this categorization may sound like a serious hindrance to any attribution of the busts to Thorvaldsen, it should nonetheless be taken with several grains of salt—for the following reasons.

To start with, Thorvaldsen himself considered his copies of classical busts to be works of lesser importance. He regarded them as practice pieces and as possible sources of income, and so did not count them as part of his true oeuvre—even though he had in fact produced them. In an 1829 interview with Thiele, for example, Thorvaldsen confirmed that he had produced the busts in Ropp’s commission, but stated that he considered them “not worth mentioning.” This issue is discussed more comprehensively in the Related Article on Ropp’s Commission; for details, please see the arguments adduced there.

Thorvaldsen’s nonchalant attitude toward these busts is the likely reason why, at the 1849 auction following his death, it was not known—or not believed—that such marble copies of classical busts could have been produced by the sculptor himself.

However, whether Thorvaldsen, late in his life, came to regard some of his earlier works as less essential is not the same as the question of whether or not he in fact had produced them. It is on the basis of this distinction that the busts are here attributed to Thorvaldsen, despite the auction catalogue’s indications to the contrary.

With respect to source criticism, the question of attribution boils down to which of the two documents we should regard as more reliable: the 1806 list—which states that Thorvaldsen had recently completed the busts—or a posthumous auction catalogue produced more than forty years after the busts’ likely formation, by people who were not present in Rome in 1806, and to whom Thorvaldsen himself can only have given the impression that the busts were of scant importance. When all is said and done, the benefit of the doubt must be given to the 1806 source—the source that is contemporaneous with the busts themselves.

An atypical choice of model?

That Thorvaldsen researchers have historically resisted attributing these two busts to the sculptor is due not merely to a lack of source material, but likely also to the fact that both the Socrates of Villa Albani and the Apollo of the Giustiniani collection are rooted in Hellenistic art. And as an artistic style, Hellenistic art has an expressive character that differs starkly from the style with which Thorvaldsen’s “original” works are otherwise generally (and automatically) associated.

It will be of benefit here to scrutinize this traditional perspective more closely. To this end, here follow a few reflections on the style in Thorvaldsen’s early Roman works:

With regard to the Socrates of the Villa Albani, for example, Helbig writes, op. cit.: “The head is excellently worked: its shapes are modeled quite plastically, and are set forth with a certain pathos …” To Neoclassicists, both expressive modeling and the explicit use of pictorial conventions of pathos seemed like remnants of the bombast of the Baroque period, which Thorvaldsen distanced himself from with his more discreet, understated formal idiom. In the case of the Apollo Giustiniani, its overwrought hairdo and dramatically turned head are characteristic of the same heroic, pathos-laden style.

Nevertheless, a case can be made that precisely these stylistic features were more prevalent in Thorvaldsen’s early work, from around the year 1800, than later. The turned head recurs, for example, in Jason, A822, 1802-03; and indeed, this more formally explicit emphasis on the drama of the Jason myth has been cited as one of the reasons why Thorvaldsen himself later came to regard his breakthrough work as inharmonious. Similarly, Thorvaldsen likely regarded his first large Homeric relief, Briseis and Achilles, A489, 1803, as all too strongly laden with drama; cf. the wrath expressed rather too directly by the Achilles figure. In his later, milder version of the same motif (A491, 1837), Thorvaldsen evidently distanced himself from this dramatic mode of expression: while the latter work is largely recomposed in the same manner as before, Achilles’ pose is notably different: he has now reined in his temper. Although this Achilles also averts his gaze in Homeric wrath, the break with the scene’s other figures is not as categorical as was the case with the earlier version of the Achilles figure. On the whole, Thorvaldsen’s second version of Briseis and Achilles can be regarded as a later attempt to integrate such a jarring element as wrath more fully into Thorvaldsen’s otherwise harmonious universe.

What is more, the first Achilles’ pose has further stylistic roots in a Hellenistic precursor—namely, a statue of one of Niobe’s fleeing sons, found in the Uffizi Gallery in Florence. Achilles’ head, meanwhile, seems to have been copied or sketched based on the heads of the two so-called Dioscuri, i.e., Castor and Pollux, the two colossal statues standing in the Piazza del Quirinale in Rome. In 1799, moreover, Thorvaldsen also made a copy of Pollux in reduced size, Dep.24, as a further attestation of his interest in the statues’ formal expression. The monumental statues in the Piazza del Quirinale present each twin as struggling with his horse; they are accordingly known as the “Horse Tamers.” This form of exertion—more broadly, of energy discharge—differs greatly from the formal idiom that manifested itself in Thorvaldsen’s later sculpture; but he evidently did not shun it completely in his earlier works.

A more monumental-heroic style is also visible in portrait busts sculpted by Thorvaldsen during the years just after 1800. In her major study of the sculptor’s portrait busts, Else Kai Sass mentions, among others, the busts of A. P. Bernstorff, 1804, A209, Princess Golitsyna, 1803-1804, A304, and Adam Gottlob Detlef Moltke, 1803-1804, A212. In the last of these portraits we find a turned head an averted gaze, which are reminiscent of that of Jason to some degree.

To the extent that it is at all possible to speak of stylistic development in Thorvaldsen’s works, one could claim that they increasingly eliminate elements that could disturb the ideal of harmony, such as turned heads and averted bodies, exaggerated plasticity and drama, and in general any pathos that is all too direct or univocal. As has here emerged, however, Thorvaldsen clearly experimented during his early career with some of these stylistic traits in various types of works.

Having thus contextualized the style of the Socrates and Apollo busts, we can now return to the question of their attribution to Thorvaldsen. As the above examples indicate, the two busts cannot simply be dismissed by arguing that their Hellenistic stylistic traits are inconsistent with Thorvaldsen’s later works. For they are not inconsistent with the earlier works here described; and what is more, there are numerous other examples of works by Thorvaldsen that flirt with Hellenistic or similar stylistic motifs. The decision to copy the Socrates and Apollo busts should thus not be regarded as an atypical stylistic choice, but rather as a reflection of the stylistic experimentation to which Thorvaldsen generally subjected his oeuvre at the start of his career in Rome. As far as style is concerned, therefore, it does lie within the bounds of plausibility to attribute these two busts to Thorvaldsen.

For an extra flourish, it may be added that two of Thorvaldsen’s most important sources of inspiration—Nicolai Abildgaard, his teacher in Copenhagen, and A. J. Carstens, his great paradigm-figure during his early years in Rome—both cultivated, in stylistic terms, a so-called “shock romantic” or Sturm und Drang form of classicism, which was greatly reminiscent of ancient Hellenism. For this reason, it is not at all surprising that Thorvaldsen’s early work should show traces of a relatively more dramatic form of pictorial expression than is found in the style to which he later subscribed consistently. Thorvaldsen, in sum, was schooled in a style quite different from the one that he ultimately distilled for his own use.


  • Arne Bruun Rasmussen Kunstauktion no. 272, Copenhagen 1971, cat. no. 714 (copy in THM 13-49/2012).
  • Auction catalogue, 1849: Fortegnelse over endeel af Thorvaldsens Værker i Marmor og Gips… m. m. af Thorvaldsens Efterladenskaber, hvilke Thorvaldsens Museums Bestyrelse lader Bortsælge ved offentlig Auction i Museets Forhalle d. 1ste Oktober 1849 og fölgende Dage, Copenhagen 1849, 88-89 (annotated copy in THM).
  • Wolfgang Helbig: Führer durch die öffentlichen Sammlungen klassischer Altertümer in Rom, vol. IV, Tübingen 1972, no. 3252.
  • Benito Vicens y Gil de Tejada: ‘De varias esculturas de Thorvaldsen que en Madrid existen’, in: El arte en España. Revista quincenal de las artes del Dibujo, Madrid, 1862, p. 286.
  • THM journal numbers 7II-2/1965 (on Socrates), 13-49/2008 (on Apollo), 13-50/2012 (on Apollo) and 13-49/2012 (on Socrates).

Last updated 27.04.2020