New light on the location of Clément Marot’s
tomb and epitaph in Turin
Article published in Studi
Francesi n. 161/2010
[anno LIV - fascicolo II - maggio-agosto 2010], 293-303
SUMMARY: That the French court poet
Clément Marot (born in Cahors in 1496) died in Turin in 1544 has always been
known; that his life-long friend Lyon Jamet wrote an epitaph for his tomb and
had it engraved in marble, also. That the epitaph was effaced and the tomb could
not be found was generally accepted. Re-combining factual information consisting
of sixteenth-century references (Audebert’s Voyage d’Italie and
Gianbernardo Miolo’s Cronaca, both not new but almost passed into oblivion), the author
reconstructs the circumstances around Marot’s burial, identifies the
commissioner, and claims that the epitaph, although effaced, can be located
almost exactly in the Cathedral San Giovanni in Turin. A recent discovery of a
contemporary drawing of the original epitaph makes a virtual reconstruction
in loco possible.

Marot's epitaph in Turin - more info at the end of this essay]
To refresh the minds of the
connoisseurs, and to introduce the others to the issue at stake, I first
present a short survey of the state of the question about Clément Marot’s
arrival and stay in Turin, trying to distinguish between known facts (F) and
probabilities (P)
-
After having left Geneva in 1543 on an unknown
date, but (F) after October the 15th, when Jean Calvin pleaded at the Geneva
council for a pension to support Marot,[1]
and (P) before 20/12/1543, when Marot’s name is mentioned but he himself is
not summoned for the Consistory in the so called tric-trac affaire.[2]
-
(F) Marot spent some time in the Savoy, i.c.
near Annécy with Bonivard’s sister in law, Pétremande de la Balme (to whom
he dedicated an epigram[3])
and with a friend of Bonivard, François Noel de Bellegarde (near Chambéry),
a man of some stature and political weight in the Savoy, to whom he also
addressed a poem.[4]
-
(P/F) Marot tried to move the King’s heart to
let him return to France. Some epigrams testify to this effort.[5]
-
(P) While in Chambéry he must have heard of
the preparations for battle and the subsequent victory of the Comte
d’Enghien (François de Bourbon) near
Ceresole (14 April 1544). (F) In an Epistle - "Epistre A Monsieur d’Anguyen"
- Marot offers his services to the
conquering hero and in an epigram he sends his best wishes to his military
camp.[6]
-
(F) He ventured south from Chambéry and (P)
via the pass of Col Mt. Cenis and the valley of Susa, he arrived in Turin,
the headquarters of the King’s governor.[7]
-
(F) On 12 September September Marot died.
-
(F) Lyon Jamet wrote an epitaph for his tomb,
which was engraved in marble and placed in the Cathedral San Giovanni in
Turin.[8]
At this point there is some
confusion, because some authorities suggest that Marot died in the
Ospedale san Giovanni Battista
and that Jamet had the epitaph inscribed
on his tomb there. The heading with which the epitaph of Jamet is
published (at least generally known since Lenglet Du Fresnoy (1731) placed it in
front of his edition of Marot’s works) is quite explicit:
Epitaphe sur le tombeau de Marot, Faict par Lyon Jamet, insculpé
en marbre en l’Eglise Saint-Jean de Turin, 1544, le 12 septembre.
The Turin cathedral is dedicated
to San Giovanni, i.c., John the Baptist.
[9]
“A la recherche du tombeau
perdu”
Of course people have searched
for traces of the tomb and the Epitaph, but in vain:

|
-
Nicolas Lenglet Du Fresnoy
(1674-1755), Marot’s first modern editor, writes in a footnote (Oeuvres vol. I,
p. xxiv, 1731) below the text of the Epitaph: “J’ai prié un de mes amis qui
alloit en Italie de voir en passant à Turin, si cet Epitaphe de Clement Marot se
trouveroit encore dans l’Eglise de St.Jean, où Lyon Jamet l’avoit fait graver.
Mais toutes les recherches ont été inutiles; soit que Marot ayant donné dans les
nouvelles opinions, on ait depuis oté cet Epitaphe, soit que le temps l’ait
effacé ou fait oublier.”
-
Georges Guiffrey
(1827-1887), the first one who tried to establish a critical edition of Marot’s
works, went to Turin himself (not so far, he was senator of the dep.
Hautes-Alpes) to look for any trace of it. This is his report (Oeuvres,
vol. I, p. 561): “Nous avons fait le voyage de Turin pour rechercher dans
l’église Saint-Jean la pierre sous laquelle devait reposer Marot; Nous n’avons
pu découvrir le moindre vestige de cette sépulture…
A défaut de
la pierre funéraire qui peut-être fut enlevée au milieu des vicissitudes de la
politique, ou dont l’inscription fut effacée par les pas des fidèles, nous avons
tenté d’interroger les archives obituaires de l’église.
Ces archives
s’arrêtent brusquement quelque temps avant la mort de Marot. Le temps a ses
caprices…”
|
|
|
Before continuing, a preliminary
question needs to be dealt with: How it is possible that a French poet with a
‘protestant’ reputation was buried in the Archbishop’s Cathedral, a poem in
French being carved out in marble and placed above the tomb? The level of
amazement can already be considerably lowered when one takes into account the
historical context: After the battle of Ceresolo Turin had become the
headquarters of the conquering army, and thus the centre of French dominion. In
1544 the term ‘protestant’ was not yet cleared out: many people were longing for
and working on a reformation of the Church; in and around Turin the Vaudois
community was prominent; the region of Piedmont was known as a safe haven for
many a refugee from France;[10]
and finally the (unifying) influence of John Calvin should not be overestimated
yet: his dominance was only emerging. One should not project (or better:
retroject) simple oppositions of later days to times when they were only
in
statu nascendi. But there is more to this than placing the facts in a
historical perspective alone.
Marot was not just ‘anybody’; he was the ‘prince des poëtes francoys’. The usual
suggestion that Marot lived and died in poverty the last months of his life in
Turin, desperately – but in vain – trying to get restored to his former glory
(the favour of the King) is not based on fact; On the contrary: the few known
facts seem to point in the opposite direction. A contemporary witness to Marot’s
burial, Gianbernardo Miolo (1506 - after 1569, since 1539 a notary employed
by Gullielmo of Cercenasco, a village near Turin), informs us that the King’s
Ambassador in Rome, George d’Armagnac
(at the end of the year he is offered the cardinal’s hat), carried the expenses
for the burial of Marot.[11]
This simple communication is revelatory, but seems to have eluded the attention
of Marot scholars. George d’Armagnac was not only the King’s ambassador, but
also Marguerite de Navarre’s protégé. She had introduced him at Court. He became
one of France’s most influential diplomats, friend of Princes and Popes, and
great lover of the Arts.[12]
He is suggested to have commissioned the burial and commanded the placement of
the epitaph, something which seems quite imaginable. Perhaps the author of the
Epitaph, Lyon Jamet, was the one who in loco
took care that everything
went as foreseen. Him we often only know as Marot’s friend, but he was much more
than that. He was seigneur de Chambrun and an international diplomate.
Ever since his flight to Ferrara, slightly preceding Marot’s arrival there, but
having fled for the same reason (he was on the list of wanted persons after the
Affaire des Placards), he was at the service of the Duchess ànd the Duke
of Ferrara,
which is quite extraordinary regarding his ‘protestant’ stigma. As the Duke’s
personal secretary and ambassador he fulfilled many an important mission, both
in Italy and France.[13]
Behind these two men, two of the most powerful female friends of Marot:
Marguerite de Navarre and Renée de Ferrara, become visible by implication. They
are the true instigators of his prominent burial place. Marot was not living an
obscure life in Turin, nor did he die unnoticed. Prominent persons took care of
his final resting place, which therefore should be worthy of France’s most
eminent poet: in the Cathedral of Turin, the Duomo San Giovanni.

|
.jpg)
|
François de Bourbon, count
of Enghien,
general of the French Army in Italy in 1544
picture: Corneille de Lyon
|
The 'Memoires' of Martin du Bellay,
brother of Jean (Cardinal) and Guillaume (Diplomat),
one of the Turin governors in 1544.
|
Nicolas Audebert’s description of the burial place and
the Epitaph
In 1962 Adalberto Olivero dug up
and published ‘new’ (i.e., once more, ‘old’) information about the exact
location of Marot’s tomb and epitaph; information he had found in a Manuscript
in London, containing Nicolas Audebert’s report of his Italian journey of
1574-1578, Voyage d’Italie.[14]
Audebert writes – with indignation – that Marot’s epitaph in Turin was erased
just before his arrival. He mentions as a matter of fact that this was
explicitly demanded by the roman-catholic authorities. Next to the very precise
date and circumstance, already noteworthy, the most interesting element of Audebert’s description is that he also indicated the exact location of tomb and
epitaph. To give the reader the opportunity to follow, we copy the transcription
as published by Olivero:[15]
Tout contre et à un
bout du palais est la principale et Cathedrale eglise nommée San Giovanni,
laquelle est très belle, grande et spatieuse.
Il y a deux entrées l'une qui est tout au
bout, et de premiee arrivée regarde droict au maistre aultel, à laquelle se
monte dix ou douze marches de pierre de taille.
L'aultre porte est petite
et à main droicte, devant laquelle y a un petit perron pour venir en l'eglise et
soubz iceluy est ensepulturé Clement Marot
duquel l'epitaphe estoit tout
proche, au dedans de l'église, en une pierre longuette qui est dans la muraille,
laquelle a[16],
depuis peu de temps,
esté martellée et l'Epitaphe effacé, par l'advis et requeste de l'Archevesque,
et rnaistres de l'inquisition, avec le consentement de son Altesse, ce que
Madame la Ducesse avoit longtemps empesché et rompu le coup quand il s'estoit
proposé.
Il n'y avoit en l'Epitaphe
qu'un dixain en vers francoys, telz qu'ilz suivent qui furent faicts par un
aultre poete francoys nommé Lyon Iamet.
Icy devant au giron de sa
mere
Gist des Francoys le Virgile
et l'Homere
Cy est[17]
couché, et repose à l'envers
Le non pareil des disans en
vers.
Cy gist celuy qui[18]
peu de terre coeuvre
Qui toute[19]
France enrichit de son oeuvre
Cy dort un mort, qui
tousjours vif sera
Tant que la France en
Francoys parlera.
Bref gist, repose, et dort en
ce lieu cy
Clement Marot de Cahors en
Quercy.
le 12 septembre 1544
Audebert renders the epitaph in
a version almost identical to already known editions. According to his own
report, he did not actually see the original epitaph, since it was already
effaced when he arrived in Turin. The information about the circumstances in
which the epitaph was demolished, seems trustworthy. One gets the impression
that Audebert reproduces information he got while in Turin. Something of fresh
felt indignation shimmers through his text. Apparently the epitaph was
demolished shortly after the death of Marguerite de France (15 September 1574),
she also being the only reason that this was not done before. The Archbishop,
Girolamo della Rovere, was educated in France, acquainted with the poets of the
Pléiade, and tried to implement the Tridentine reform. The consent of the
Duke of Savoy can also be understood. He not only supported the new archbishop,
but seemed to have had high expectations of the Jesuits:[20]
signals that both opted for a re-catholisation of the Waldensian region and
therefore were willing to cooperate with the ‘masters of the inquisition’ to
erase Marot’s epitaph in the Cathedral. Marot’s ‘fama lutherani’, during his
lifetime already inextricably bound to his person, had only increased after his
death, in particular because his Psalms were sung in reformed liturgy. Next to
these religious motives, one should also not underestimate anti-French
sentiments in Turin/Savoy in those years. The French occupation (from 1536) had
ended in 1559, when the duke of Savoy (Emanuele Filiberto) had succeeded in
transforming his duchy into a powerful political player in the region (peace of
Cateau-Cambrésis). Italian became the official language and in the centuries to
come the Duchy of Savoy became a stable and unifying factor on the hopelessly
divided Italian peninsula. This relative independence (both from Spain and
France) of the Duchy of Savoy coincided with the Duke’s marriage with the
daughter of François I, Marguerite de France. The personal attachment of
Marguerite de France (1523-1574) to the ‘monumentum’ for Clément Marot can also
be understood: She must have known him personally, Marot was her father’s
official poet; as a young girl she even once had received an Epistle from her
niece (Jeanne d’Albret), which in reality was written by Marot.[21]
As a girl from François’s first marriage (with Claude de France, d. 1524), she
was raised by her aunt, Marguerite d’Alençon, Queen of Navarre, Marot’s most
loyal supporter and protector. Cultural interest, spiritual open-mindedness, and
readiness to personally protect religious refugees, mirror her upbringing. This
valuable information provided by Audebert, made available by Olivero in his
article in 1962, did not really attract attention of the scholars in het last
part of the twentieth century.
In his biography of Marot (1972),
Mayer obscures this when he, in a footnote referring to the article of A.
Olivero (!),
only writes: “Peu de temps après l’Inquisition semble avoir enlevé
toutes traces de ce tombeau. Déjà au dix-huitième siècle il était introuvable.”[22]
Location of Tomb and Epitaph in the Duomo
But what is even more
astonishing, is that the very precise indications about the location of Marot’s
final resting place and the Epitaph inside the Church seems completely to have
eluded the eyes of modern scholars, since I could not find any reference to this
in subsequent literature. Nevertheless it can’t be more precise. Audebert gives
accurate directions, as if he wants to guide the reader to the proper place. To
find Marot’s grave the visitor should not enter through the main entrance but
take the smaller door at the right-hand side of the Church.
-
In front of this door is a landing, pavement
(‘perron’). Here Marot is buried (“et soubz iceluy est ensepulturé Clement
Marot”)
-
To find Jamet’s epitaph one should enter the
church through that door, and look for the epitaph, since it should be
nearby (“duquel l'Epitaphe estoit tout proche au dedans de l'église”);
-
It should not to be looked for on the floor
(as Guiffrey did), but on the wall (“en une pierre longuette qui est dans la
muraille”
-
One should not expect to find it, since it was
completely demolished and cut off (“martellée”). Nevertheless the location
might still be determined.
Based on this information an
“expedition” to Turin forced itself. Although no expert in architecture and
inscriptions at all, and with only a very general knowledge of the history of
Turin, this could never be more than prospecting to size up the situation and
determine whether a further investigation would be worthwhile. The results – as
described below – I offer to real experts ‘as they are’, i.e., without any
pretension, hoping they might incite them to make the proper assessments in
loco.[23]
The ‘small entrance at the right-hand side of the Cathedral’ was quickly found.
The space in front of it, where the ‘perron’ used to be, serves as an office to
the parish of San Giovanni. The space below, where Marot originally was buried,
is now used as the entrance of the museum. The many changes, restorations and
transformations made it highly unlikely that the bones of Marot would still be
there, but the location of the burial place itself seems certified.
 |
 |
1a. outside view |
1b. inside view |
Ever since the remnants of the
old churches were discovered under the existing church (restoration of 1997,
after the great fire), archeologists and architects have taken over to uncover
and interpret the presence of ‘a complete second church’ below the ‘upper
church’, apparently also meant for devotional use. Next to the remnants of three
palaeo-christian churches, they found and identified the bones of Cardinal
Domenico della Rovere
(d. 1501), bishop of Turin and driving force behind
the construction of the Renaissance cathedral completed in 1498.[24]
These excavations brought to light that not only the crypt below the sacristy (a
little further at the right side of the choir) was used as an ossuary: burial
places were found all along the outside church walls.[25]
One is still in the process of making the inventory.
 |
2. profile - longitudinal section –
of the Duomo (with the contours of the chapel of the Holy Shroud, a
later addition). The entrance is between the 6th and 7th
pillar, just before the transept. |
It is apparent that in the early
days of this Cathedral this area was used for burials, elements not only
corroborating the account of Audebert about the ‘perron, soubz iceluy est
ensepulturé Clement Marot’, but also providing it with the necessary context to
make it imaginable.
The church itself is full with
epitaphs and funerary monuments, many of them placed on the wall, engraved in
marble. According to Audebert’s report Marot’s epitaph was located inside the
church, on a stone in the wall, not far from the door (‘tout proche au dedans de
l'église en une pierre longuette qui est dans la muraille’). Both walls close to
the door were equipped with inscriptions. To the right (when entering the
church) there is an inscription, beautifully carved in marble with an elevated
border. Above the text is the coat of arms of a noble family (the inner part is
vanished, only the outside shape is visible). It must belong to the marquisat of
Ceva, since according to the Epitaph a marquis of Ceva, named Cristoforo,
was buried there (‘Christophorus marchio Cevae’).[26]
 |
 |
3. Epitaph of Cristoforo di Ceva (at the
right-hand side from the door on entering the Church) |
3b. The entire stone with the impression
of depth (no flash, therefore unsharp). |
The reason why he was buried in
the Turin Cathedral is also mentioned: He was related to (‘nepos’) Domenico
della Rovere himself, in the text simply referred to as the cardinal of S.
Clemente (‘cardine[us] Sancti Clementi’). Cristoforo di Ceva died 15 May
1516. We can assume that the stone was fix and firm at that particular place
when in 1544 Jamet’s epitaph for Marot was to be placed on a wall near the door.
On the opposite wall (left when
entering the church) a stone remembered that that Claude Guichard was
buried there, counselor and historiographer of the Duke of Savoy, a famous
archeologist (specialized in ancient funeral rites) and French poet as well. The
epitaph informs us that he had died 8 May 1607.[27]
 |
4. Epitaph of Claude
Guichard (at the left-hand side from the door on entering the
Church) |
This epitaph postdates the
removal of Jamet’s inscription with 33 years. It therefore is quite possible
that this was the wall ‘tout proche’, on which until 1574 Jamet’s Epitaph could
be read. We took a closer look and noticed something odd. Contrary to the
epitaph on the opposite wall (and many others in this church), this epitaph was
not carved out beautifully in marble, with a border as so many others in the
church; it appeared to be not even really engraved: it was more painted
on the stone than carved in it. A rectangular space in a whitened wall. Taking
an even closer look, we noticed roughness at the place where the upper and lower
border (but it was no border, the white painting simply stopped there), as if
something had been cut off and the surface was not properly smoothed. Of course
the church has been damaged, restored and repainted many times. And perhaps
there is another explanation for this peculiarity, but nevertheless: looking at
this post-Audebert inscription, it seemed quite imaginable that his inscription
covered the place of a previous one, that of Jamet’s epitaph for Marot, which
had been removed by force (“martellée”).
[28] All elements fitted:
not far from the right side door…
just inside
the church…
Although we had not seen Jamet’s
epitaph with our physical eyes, we had the strong impression of having seen it
with our spiritual eyes, a minor but sweet revenge on those people who had so
vigorously tried to wipe out all traces of Clément Marot de Cahors en Quercy.
La mort n’y
mord
Antwerp, 25 July
2009 // Dick Wursten & Jetty Janssen
Post Scriptum
After
returning home, having done the research to write this article, our attention
was drawn to a recent publication by Richard Cooper, in which he relates that he
found an image of the original epitaph in Turin, a drawing by a student, who in
the middle of the sixteenth century travelled through Italy, Spain, Germany and
France setting down epitaphs and other inscriptions. His Manuscript ends with
two epitaphs dedicated to Marot, the final one being Jamet’s.
(Harvard,
Houghton Library, MS Typ 152, f° 179v°,
Imagines
sepulcrorum et epitaphiorum inscriptiones antiquae.) See, Richard Cooper,
‘Dolet et Marot jugés par Jean Binet et Gabriele Siméoni’, in Esculape et
Dionysos: Mélanges en l’honneur de Jean Céard, ed. Jean Duprèbe, Franco
Giacone, Emmanuel Naya (Geneva, 2008), p. 511-527, the discovery on p. 522-523,
the reproduction at p. 527. The text of the epitaph in majusculs with
abbreviations and orthographical errors (which at the same time betray that the
one who carved it probably was an Italian and that the drawing is not fake)
closes with: “Obit Thaurini An.D. M.D.XLIIII. D.XII..S.” (‘he died Anno Domini
1544 on the 12th Day of September’). The space used for Guichard’s
epitaph and the space needed for Marot’s seem to match.
The drawing of this
wandering student is reproduced heading this article, the entire page below (copy from the article of Richard Cooper,
p. 527).

“Maistre Calvin pour Clement Marotz. – Le sieur Calvin a exposé pour et
au nom de Clement Marotz requerant luy faire quelque bien et ilz se
perforera [usually emended: se parforcera] de amplir les seaulme
de David. Ordonné de luy dire que pregnent passience pour le present.”
(P. Pidoux, Le Psautier Huguenot, vol. II (Basle, 1962), p. 23.
The story can be read in the minutes of the Consistoire from 18 &
20 December 1543 (Registres du Consistoire de Genève,
vol.
I, p. 287‑295). Also extracts in Pidoux II, p. 23‑24). The games that
are mentioned are dames, jeux des cartes, dés,
and
tricquetract and the core issue is to identify the ‘prédicant’ from
Orbaz (Pidoux transcribed ‘Orléans’) who is suggested to have
participated in a game played for money. Reading between the lines, one
can deduce that François Bonivard (Seigneur of St. Victor), Clément
Marot and Aimée Curtet (a syndic) met regularly to play.
“Adieu ce bel oeil tant humain…” (Clément Marot, Oeuvres Poétiques,vol.
II, ed. G. Defaux, p. 337‑338).
A ung sien amy (Defaux II, p. 703‑705). We do not take into
account the anonymous epistle addressed to M. Pelisson XE "Epistre A
Monsieur Pelisson, President de Savoye*" (president of the
Parlement of Chambéry):
A Monsieur Pelisson, president de Savoye,
1543, (Defaux II, p. 705‑707). The author of this epistle has only
recently arrived in Savoy coming from Paris/France (v. 35). The implicit
chronology is irreconcilable with established elements of Marot’s
chronology, who left France in 1542, stayed in Geneva for about a year.
Discussion of the authenticity in Mayer, vol. 1 [Les Epîtres], p.
62‑63 (rejection), and in Defaux II, p. 1292‑1294, cf. p. 1122
(acceptance of the authenticity, but not successful in accounting for
the chronological problems implied). That the epistle is too feebly
constructed and old-fashioned to have been written by Marot, is perhaps
not a cogent argument contra (Defaux makes this point against Mayer),
but it has some relevance; it at least does not counterbalance the
simple bibliographical and text-internal argumentation as presented by
C.A. Mayer (following Pierre Villey).
A ‘placet’: “Plaise au roy congé me donner / D’aller faire le tiers
d’Ovide…" (Defaux II, p. 710). ‘Le tiers d’Ovide’ refers to the
translation of the Metamorphoses by Marot; a
Dizain au Roy.
envoyé de Savoye. 1543 (Defaux II, p. 319; first publication 1547
(Marnef); also present in Du Moulin and Fontaine’s selection of Marot’s
Oeuvres, so bibliographically as ‘certain’ as possible).
This epistle was published separately in 1544 by N. L’Heritier (Mayer,
n° 121, Defaux II, p. 707). The epigram Salutation du camp de
Monsieur d’Anguiers appeared in print in 1549 in an edition of
Marot’s Oeuvres by Jean de Tournes (Mayer n° 169, Defaux II, p.
338). The style and content of this epigram leave open the possibility
that Marot had already joined the camp before the actual battle.
The final phrase in l’Histoire ecclésiastique
concerning Marot
suggests the same: “…il s’en alla passer le reste de ses jours en
Piémont, alors possédé par le roy, où il usa sa vie en quelque seureté
sous la faveur des gouverneurs.” (Histoire Ecclésiastique des Eglises
Réformées, vol. 1, ed. P. Vesson (Toulouse, 1882), p. 20; reprint
from the 1580 edition, ascribed to Th. de Bèze. One of the actual
governors at that time was Martin du Bellay, brother of Guillaume, who had
been governing the province of Piedmont until his death in 1543.
The epitaph was published in Cinquante Deux Psaumes de David
(Paris, Jacques Bogarde, 1546), Mayer n° 149. Already on 1 October 1544
(date of the privilege) a Déploration de France sur la mort de
Clément Marot, son souverain poëte, appeared in print in Paris. See
Guiffrey I, p. 563. For the text of this epitaph, see Mayer Clément
Marot , p. 514-515 (idem in Defaux II, p. 1187). Guiffrey I gives
the text as published by Roville 1561.
An Ospedale San Giovanni indeed existed. It was even officially
(re-)instituted in 1541in an effort to improve social welfare by
centralising all city hospitals in a modern building near the Duomo
(Sandra Cavallo, Charity and Power in Early Modern Italy -
Benefactors and their motives in Turin, 1541-1789, [Cambridge
History of Medecin] (Cambridge, 1995), p. 12-14. The hospital only
received a proper building near the cathedral in 1545.
The first occurrence of Marot being buried in the
(chapel of) the hospital, I found – but I don’t claim completeness – in
1914 with Ph.A. Becker in the final part of his biographical essay
‘Marots Leben’,
Zeitschrift für französische
Sprache und Litteratur 42 (1914), p. 205:
“Marots Leiche wurde im Ospedale San Giovanni Battista in Turin
beigesetzt, und sein alter Freund Lion Jamet setzte ihm die
Grabschrift.” In 1926 (in the introduction to his biography) Becker
himself classified this text as premature, though leaving this error
uncorrected (Clement Marot, sein Leben und seine Dichtung, p.
183). The same error I also found with Henri Guy (1926, Clément Marot
et son école, p. 320 ); Pierre Jourda (1950/1967,
Clément Marot,
p. 58) ; C.A. Mayer (1972, Clément Marot, p. 515, and G. Defaux
(1990, Oeuvres Poétiques t. I, p. clxvii). The fatality of such a
copying attitude is that the misinformation is raised to the status of
fact. E.g., as such it is presented in the popular biography by Jean-Luc
Déjean, Clément Marot (Paris, 1990), p. 383-384 : “Ce dont nous
sommes sûrs, c'est qu'il [sc. Jamet] arriva à point pour lui donner des
funérailles décentes, dans le cimetière de l'hôpital Saint-Jean-Baptiste
à Turin. Il orna la tombe d'une plaque de marbre portant une épitaphe de
sa façon.”
See J. Jalla, ‘Le refuge français dans les vallées Vaudoises’, in BSHPF 83 (1934), p. 561-592 ;
BSHPF 85 (1936), p. 5-25. He
writes about the beginning of the 1540s : "La tolérance religieuse était
tellement plus grande en Piémont que dans les autres provinces de la
monarchie française, que cette région servit de refuge à plusieurs de
ceux qui étaient traqués au-delà des Alpes." (p. 573). For a general
assessment of the Vaudois movement, id. Storia della Riforme in
Piemonte (Torre Pellice, 1914; reprinted in 1982).
Cronaca di Gianbernardo Miolo di Lombrascio notaio. The notes
concerning the events that happened during his lifetime, were extracted
from his manuscript by Giuseppe Vernazza and prepared for print in 1771.
Vernazza comments on Miolo’s style and notes: “Lo stile adoperato dal
Miolo è rozzo latino, ma dappertutto risplende la buona fede e la
natural franchezza della verità.” These ‘Notizie’ were published in
Miscellanea Storia Italiana, vol. 1, ed. Regia deputazione di Storia
Patria (Turin, 1862), p. 145-233. The quote about Marot’s funeral is
placed at the end of a detailed account of the main events of 1544 of
the region (including the battle at Ceresole). Miolo apparently did not
know the exact dates, since normally his chronicle is full of exact
dates ordered chronologically. As a kind of Post Scriptum to the events
of 1544, he writes: “De anno eodem 1544. Taurini Clement Marot gallus in
rittimis galicis clarissimus moritus. et in templo archiepiscopali
inhumatur expensis Georgii cardinalis Armeniaci.” (p. 184). Jalla, a.c.
p. 576 (see above note 10)
and A. Olivero in his edition of Nicolas Audebert’s Voyage d’Italie,
(Rome, 1981), p. 291 both refer to Miolo’s chronicle.
George d’Armagnac (1501-1585) was portrayed by Tiziano Vecelli (now in
Musée du Louvre). For him, see Salvador Miranda, The Cardinals of the
Holy Roman Church, Essay of a General List of Cardinals (112-2007),
http://www.fiu.edu/~mirandas/bios1544.htm#Armagnac.
On Lyon Jamet, see Rosanna Gorris,‘"Va lettre, va ... droict à Clément’:
Lyon Jamet, sieur de Chambrun, du Poitou à la ville des Este, un
itinéraire religieux et existentiel’, in Les Grands Jours de Rabelais
en Poitou. [Actes du Colloque de Poitiers réunis par Marie-Luce
Demonet] (Geneva, 2006), p. 145-172. Jamet lived in Ferrara from
1535/6-1548, returning there once more in 1554 to assist Renée when she
was imprisoned (during an investigation of the Inquisition).
British Museum, Lansdowne Ms. 720. This Ms. has been ‘known’ always, but
only in the last part of the nineteenth century researchers began to pay
proper attention to it (Richter, Müntz, De Nolhac, Picot). Nicolas
Audebert travelled through Italy with recommendation letters from
influential friends and Latin poems from his father (Germain Audebert,
who had done a trip through Italy in 1539). Nicolas arrived in Turin on
22 October 1574 and left on 27 October. Adalberto Olivero, ‘Una
testimonianza trascurata sulla tomba di Clément Marot a Torino’, Studi Francesi 16 (1962), p. 263-265. A critical edition of Nicolas
Audebert, Voyage d'ltalie, appeared in two volumes, procured by
the same Adalberto Olivero (Rome, 1981), with extensive introduction,
bibliography and footnotes.
Olivero, ‘Una testimonianza trascurata…’, p. 264. In the critical
edition of Audebert, Voyage d'Italie, p. 141. Original: Ms.
Lansdowne 720, f. 37 r°/v°. We noticed some minor differences in
capitalisation and interpunction, and one corrected error (. Since
Olivero signaled that Jamet’s verse is written in a larger hand than the
rest of the text, we did the same in print. The text of the Epitaph
(except orthographical differences) is identical with the text of
Roville 1561 (Guiffrey/Douen); in the 1962 article. The transcription of
1981 differs in verse 5: “qui” instead of “que” (1962). Partial
translation: “…At the right-hand side of the church is a small door, in
front of which there is a pavement (or ‘landing’. French: ‘perron’) to
get into the church and beneath this pavement Clément Marot is
buried; his epitaph is nearby inside the Church, on a oblong
(rectangular) stone in the wall, which has only very recently been
chopped off and the Epitaph erased, on the explicit advice and request
of the Archbishop, and the masters of the Inquisition, with the approval
of his Highness, something which Milady the Duchess for a long time had
precluded, obstructing the plan when it was proposed…”
“qui
” (Bogarde 1546); “que” (Roville 1561).
“toute la” (Bogarde 1546); “toute” (Roville 1561).
Since 1560 the
Jesuit missionary
Antoine Possevino actively tried to re-catholicise the Waldensian
valleys. When the Duke restored the Turin University he recruted
professors of theology and metaphysics among the clergy and decreed that
members from the newly founded (1567) Jesuit college would teach the
humanities. (Paul F. Grendler, The universities of the Italian
Renaissance, p. 88-89). That the Duke organised the transfer of the
Holy Shroud from Chambéry to Turin in 1578, also speaks volumes, even
more because it apparently was staged to please the very pious bishop of
Milan, Carlo Borromeo.
Pour la petite princesse de Navarre, à Madame Marguerite, “Voyant
que la Royne ma Mere…” (Defaux I, p. 330). Jeanne d’Albret (ghost)writes
this letter to reassure her niece. The famous ‘Ma mignonne / Je vous
donne / Le bon jour…’ is meant to cheer up the same patient. For
Emanuele Filiberto, see P. Merlin, Emanuele Filiberto, un principe
tra il Piemonte e l’Europa (Turin, 1995) and for Marguerite de
France (or of Savoie), see R. Peyre, Une princesse de la Renaissance,
Marguerite de France, duchesse de Savoie, (Paris, 1902), awaiting a
new monography by Rosanna Gorris, Marguerite de France, princesse des
frontières. Poésie, éthique et politique à la cour de la duchesse de
Savoie (forthcoming).
Mayer, Clément Marot, p. 515. With this last phrase Mayer
probably refers to Lenglet’s initiative as does O. Douen in his book
about Marot and the Huguenot Psalter (vol. I, p. 443, footnote): “On a
vainement cherché son épitaphe dans l’église Saint-Jean, en 1731; il est
probable qu’elle avait disparu dès le
xvie siècle.” A
short summary of Olivero’s article, focusing only on the circumstances,
not mentioning the location, was published by H.P. Clive in his research
bibliography Clément Marot, an annotated bibliography (London,
1983), sub C 111.
We visited Turin on Monday 13 July 2009. The museum was closed (only
open on Friday-Saturday-Sunday) and we only had time to take a quick
look around and make some pictures.
The
Della Rovere family was “one of early modern Italy's most powerful and
influential historical families… a family of popes, cardinals, and
powerful dukes who financed some of the world's best known and greatest
artwork.” (back-cover of a collection of essays entitled: Patronage
and dynasty: the rise of the della Rovere in Renaissance Italy, ed.
Ian F. Verstegen (Truman State University, 2007). The most famous Della
Rovere pope is Julius II (Giuliano Della Rovere, pope from 1503-1513,
father of Felicia). Domenico became cardinal in succession of his older
brother Cristoforo, who in 1478 was made cardinal of Tarentaise by pope
Sixtus IV (Francesco Della Rovere, no direct family relation). Domenico
preferred the title of S. Clemente in 1479; he was transferred to Geneva
in 1482, but exchanged it for the see of Turin in the same year. In Rome
he acquired a chapel in the church of S. Maria del Popolo, which he had
Pintoricchio decorate. At his death in 1501 he was buried there, next to
his brother. Domenico’s remains were later transferred to Turin and
buried in the crypt of ‘his’ cathedral.
The bones of Anne de Crequi, (died 1541), the wife of
Sieur de Langey (Guillaume du Bellay, governor of Piedmont) were also
found and identified.
Next to the neatly labelled cases with the bones of Della Rovere and
other clerics in the crypt below the sacristy “[i] sotterranei del Duomo
custodivano altri tipi di sepolcri. Numerose lapidi, indicazioni sui
muri e incisioni attendono di essere studiate e messe a confronto della
storia. Oltre all’ossario degli ecclesiastici sono state rinvenute a
ridosso delle mura perimetrali lunghe sequenze di tombe a botola.”
Alberto Ricadonna, ‘Ecco l’ « altro » Duomo. Un magnifico complesso
liturgico e archeologico – La tombe di Della Rovere’ in La Voce del
Populo,
http://www.diocesi.torino.it/exdiario/altro_duomo.htm (last
modified, 10 September 2003; accessed 23 july 2009). This article was
published on the occasion of the completion of the restoration of the
crypt/ souterrain of the Cathedral. The supervising architect was
Maurizio Momo. It became clear that Cardinal Della Rovere had created
two churches, one upper and one lower church. A museum was instituted by
the diocese to give the people access to the souterrain of the Cathedral
and the discoveries made there. M. Momo wrote an official text to
introduce this museum: ‘Il Museo Diocesano. sede – restauri –
allestimento’.
http://www.diocesi.torino.it/museo/scheda.htm (last modified 13
January 2009; accessed 23 july 2009). The plans and photos are taken
from an article about the Museum by Don Natale Maffioli (in a brochure
in which Gianluca Popolla describes the way an ecclesiastical museum
should function), apparently meant to instruct the volunteers (‘corso
per volontari del museo diocesano di Torino’):
http://www.diocesi.torino.it/museo/volontari-museo-marzo09.pdf (last
modified 19 March 2009, accessed 23 July 2009).
Ceva is one of the many marquisats in Piedmont, itself divided in a
number of minor marquisats. It losts its independence (i.e. direct
feudal link with the Emperor) when Charles V granted it to the Duke of
Savoy in 1531. The text (entirely in majusculs): “hoc tvmvlo rari
splendoris dona ferv[n]tvr / hic e[st] christophorvs tvmvlatvs marchio
cevae / Cardineiqve nepos patris cognomine sancti / Cleme[n]tis sacri
templi reverendvs et hvivs / Canonicvs qvovis censendvs honore sacerdos
/ Moribvs: ingenio vita: probitate: decore. / Obiit xv maii m.d.xvi.”
The “a” from cevae is not placed correctely and twice a small majuscul
is needed to correct a mistake (“marchio” and “cardineique”). Standard
abbreviations are used. I was not able to identify a Cristoforo as
marquis of Ceva. For the history of Ceva, see Giovanni Olivero, Memorie storiche della città e marchesato di Ceva (Ceva, 1858). He
identifies this person as the son of Aria di Valarano della Rovere,
sister of the Domenico della Rovere, and Gio. Antonio Ceva d’Ormea. (o.c.,
p. 122. See also Ferdinando Rondolino, Il Duomo di Torino illustrato
(Turin, 1898), p. 171.
This epitaph is also carved entirely in majusculs. Abbreviations are
used, once with minor uppercase letters in superscript (SERMI]:
“Clavdivs gvichardus ara[n]dati dominus / ab intimis consilijs
svpplicibvsqve / libellis ser[enissi]mi, sabavdiæ dvcis hic / post
varios casvs ad / aeternam qvietem / qvuescit. / Soli fide deo vitae,
quod svfficit opta. / sit tibi cara salvs, caetera crede nihil. / vixit
annos li, dies xxix. / obiit die viii. maij: / m.d.c.vii”. Claude
Guichard (born around 1545), studied in Turin, was a close friend and
colleague of the (more famous) Antoine Favre. Guichard is mainly
remembered for his Funerailles et diverses Manieres d’ensevelir des
Rommains, Grecs et autres nations (Lyon, Jean de Tournes, 1581),
dedicated to the Duke of Savoy. He also translated Livius and published
Quatrains sur la vanité du monde, a poetic genre very popular at
that time in France. Intriguing is the adagium at the end of this
epitaph (“Soli fide Deo vitae quod sufficit opta / Sit tibi cara salus
caetera crede nihil.”: ‘Trust in God alone, desire from life what is
sufficient, Take good care of your salvation, for the rest fear
nothing’), not so much for the superficial resemblance to ‘sola fide’
but because it is quoted 50 years later by Guy Autret, Seigneur de
Missirien, in the dedication to the Bishop of Cornwall of his 1659
edition of Albert Le Grands Vie des Saints de la Bretagne. He
refers to this adagium as written by “un auteur pieux”, caracterising it
as one of the finest summaries of what faith is about ever written. (p.
xxii, edition Quimper, 1901). This suggests that it was published. We
wonder, did Guichard write it, publish it? or his friend Favre? or…?
The next epitaph (also on the left side of the door, but on the marble
pillar, just visible on fig. 1b) was carved out in an beautiful manner
as well including an ornamental layer. It is the epitaph of “Anto.
Adimarus” a famous Florentine (Antonio degli Adimari), who died 1528,
not only farther away from the door than the epitaph of Guichard, but
pre-dating Marot’s death as well.
|