Economic History of Opera François R. Velde * Federal Reserve Bank of Chicago July 31, 2015 Abstract Using a database of performances at five major opera houses from c1750 to c1950 I document that a shift in the performance of operas occurred in the mid-19th century, away from performing only recent works and for a short while, toward the emergence of a growing canon of known works and composers. The shift occurs roughly at the same time in all the theaters, suggesting that continent-wide factors were at play rather than national idiosyncracies. Looking for economic forces to explain the shift, I document the financial history of one theater, the Paris Opéra, from 1803 to 1914. Like many other theaters at the time, it was mostly managed by for-profit entrepreneurs. Problems of rising costs and high risks were constant preoccupations. The main rise in costs appears to be in the wages of artists, particularly soloists, with increasing inequality and the emergence of a star system. A strategy of risk-avoidance may have driven directors toward programming works with known track record of success. Apologies to EHES 2015 participants for this very rough, preliminary, and incomplete draft. Keywords: (JEL ). *
Economic History of Opera
Mar 15, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Economic History of OperaJuly 31, 2015
Abstract
Using a database of performances at five major opera houses from c1750 to c1950 I document that a shift in the performance of operas occurred in the mid-19th century, away from performing only recent works and for a short while, toward the emergence of a growing canon of known works and composers. The shift occurs roughly at the same time in all the theaters, suggesting that continent-wide factors were at play rather than national idiosyncracies.
Looking for economic forces to explain the shift, I document the financial history of one theater, the Paris Opéra, from 1803 to 1914. Like many other theaters at the time, it was mostly managed by for-profit entrepreneurs. Problems of rising costs and high risks were constant preoccupations. The main rise in costs appears to be in the wages of artists, particularly soloists, with increasing inequality and the emergence of a star system. A strategy of risk-avoidance may have driven directors toward programming works with known track record of success.
Apologies to EHES 2015 participants for this very rough, preliminary, and incomplete draft.
Keywords: (JEL ).
1 Introduction
This paper is a first venture into the economic history of opera. The first point is that a substantial shift in the performance of operas took
place in the middle of the 19th century. Before then, operas were produced in relatively large quantities and performed for a few years before being forgotten. There were a few exceptions, mostly local (Paris tended to have a culture of classics more than other opera houses; Vienna tended to preserve the memory of Mozart when he was forgotten everywhere else).
Can this be explained by economic factors? Of course, it is possible to ascribe this to preferences, or intrinsic quality. The latter explanation is simple: until 1850, no works worth keeping were produced. This is a little problematic: not only were good works produced before 1850, in fact a substantial part of the modern repertoire is made up of those works, rescued from oblivion over the years. Preferences might explain the fact, as follows: until the mid-1850, opera-goers valued novelty over quality. After 1850, tastes shifted toward quality. This explanation is only partial (why the shift?) and also does not explain why new works were constantly added to the repertoire after 1850.
My approach is to look for a documented shift in factors external to the opera industry. The intuition (yet to be modeled) is that increasingly high fixed costs made the performance of new works financially risky, and drove risk-averse entrepreneurs toward the safety of known successes.
To document this I turn to the Paris Opéra, not perhaps the most represen- tative institution but the one I could research relatively easily. From 1831 until the 1930s it was managed by for-profit entrepreneurs who leased the institution. Admittedly they operated under many constraints which I document. They also received a sizeable subsidy but, as I show, it was lump-sum and hence did not affect decisions at the margin. I reconstruct annual budgets from 1803 to 1914 and look at revenues and costs, as well as capacity and pricing. I focus particularly on salaries of artists, which was the main source of growth in costs until the 1880s (hence during the period of the shift I wish to explain).
1
Tentative speculations will be offered in lieu of a conclusion.
2 The Formation of the Operatic Canon
2.1 The Repertoires of European Cities
The starting point of the enquiry is a database of all operas performed in the major European opera houses:
• Paris: Opéra (1750–1945) • Paris: Opéra-Comique (1826–1914) • Naples: San Carlo (1738–1987) • Milan: La Scala (1778–1963) • London: King’s Theater – Covent Garden (1760–1957) • Vienna: Burgtheater / Kärntnerthortheater – Hofoper (1770–1944)
There are in fact two databases: one containing the identity and count of each opera performed in each year and each theater, and another database of operas with their characteristics: composer name and biographical information (birth and death date and place) and date and place of the first world performance. I then use the two databases to study the characteristics of the repertoire (in one sense of the word, namely, what is being performed in a given place at a given time).
Figure 1 displays the basic pattern. It plots the average age of the operas performed in each theater. The age of a work is the time elapsed since its first world performance, and the ages are weighted by the number of performances in each year. A very clear pattern emerges, with two regimes. Before the mid-19th century, the average age in most theaters is between 0 and 10. Paris is a bit of an outlier: one can discern periods of rising average age, or aging repertoire, one peaking in 1770 (shortly before the works of Gluck replaced those of Rameau), and another peaking in the 1820s. From the mid-19th century, the average age begins a steady rise everywhere, with a pause in the late 19th century when new works are being added, but then resumes and continues at a 45-degree angle through the 20th century.
This may seem surprising to those familiar with the modern repertoire of opera houses: there was a time when operas were performed as soon as they were composed, played for a few years, and then tossed out. Opera, even of
2
10
20
30
40
50
60
70
80
90
Comique La Scala London Paris San Carlo Wien
Figure 1: Average age of operas performed in major European opera houses.
the high-brow sort, was entertainment rather than culture, more like cinema today. Figure 1 dates the beginning of opera as canon to the mid-19th century.
The average is only one summary statistic. Many others can be computed. Figures 2 and 3 show the quantiles of the distribution. An important element of the pattern of aging becomes clear: new works are continuously being added, and some do join the repertoire, tending to lower the average age. But at the same time old works are also being resurrected and added to the repertoire, tending to counteract this lowering. Some vintages appear to contain more durable works, tracing parallel 45-degree lines in the age distribution.
Figure 4 shows the following statistic: in each theater and each 5-year period, I consider the works premiered, tally up their cumulative number of performances over time for those works, and show the average per cohort. a flat line means works from that half-decade are not performed, a rising line means works from that cohort are being performed. Some lines rise for a long time and then stop, indicating the emergence of “premature canons.”
Are the composers being performed dead or alive? Figure 5 shows the answer. The same broad pattern: there is a big difference between pre-1830 and post-1870, although again Paris is an outlier.
3
50
100
50
100
50
100
50
100
50
100
50
100
150 Wien
Figure 2: 25th, 50th, and 75th percentiles of the age of operas performed in major European opera houses. Age is on the vertical axis.
Although the pattern of aging is broadly similar across countries, there are important differences. One emerges when we consider the origin of works performed. Figure 6 shows the proportion of works performed in each theater that were premiered in that theater, that city, or that country. For this exercise contemporary rather than modern borders are used to defined countries, except for Italy, in which I included Lombardy (where Milan is located) and the Veneto (where Venice is located) even before their annexations in 1860 and 1866 respectively.
There is a pattern of internationalization of the repertoire through the 19th century, but it proceeds at different paces in the various cities, with London and Vienna being more open to foreign works earlier. Paris is the most closed city, with both the Paris Opera and the Opéra-Comique displaying a stronger home bias for longer.
4
50
100
50
100
50
100
50
100
50
100
50
100
150 Wien
Figure 3: 10-% quantiles of the age of operas performed in major European opera houses; age is on the vertical axis. The read line is the median.
2.2 Composer Productivity
A third database, which contains essentially all operas ever performed (some- where around 33,000 distinct works) allows me to point out another interesting pattern, on the supply side so to speak.
A big shift starts in 1800, and is completed by 1850: • the number of active composers doubles (“active” means he has already
premiered his first work, not yet his last) • composer productivity falls by half before the emergence of the canon • the time elapsed between premieres doubles in the same period • this is not sensitive to exclusion of lighter works or exclusion of “amateurs” (composers with only one work)
2.3 A hypothesis
As an economist, I look for economic explanations. I also share my tribe’s prejudice that preference-based explanations are, if not the last refuge of
5
50
100
50
100
50
100
150
200
200
400
600
50
100
100
200
300
400
Figure 4: Average cumulative performances by 5-year cohorts.
scoundrels, a deeply unsatisfying route. This doesn’t mean I won’t model demand, and accept the notion of a quantity-quality tradeoff. But I wish to keep preferences constant.
Nor am I interested in the “objective quality” view. As a lover of Monteverdi, Cavalli, Lully, Haendel, Rameau, Mozart, and many other pre-Verdi composers, I will not postulate that works produced after Il Trovatore were simply better.1
What I wish to explore is an exogenous shift in costs, which, combined with an undiversified, risk-averse entrepreneur, will bias programming choices toward works that are known to have a good track record, and avoid risky ventures on unproven works or composers.
The rest of this paper is a “deep dive” into the financial history of one institution, the Paris Opéra, over a century, a period that spans the shift that I wish to explain.2
1Sorry, Joel Mokyr. We’ll just have to disagree. 2Figures 7–9 suggest that something interesting is also happening upstream, on the production side.
6
1760 1780 1800 1820 1840 1860 1880 1900 1920 1940 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Comique La Scala London Paris San Carlo Wien
Figure 5: Share of dead composers being performed.
7
0.6
0.8
0.2
0.4
0.6
0.8
0.2
0.4
0.6
0.2
0.4
0.6
0.8
0.2
0.4
0.6
0.8
0.2
0.4
0.6
0.8 Wien
Figure 6: Share of works locally premiered, by theater, city, and country.
1650 1700 1750 1800 1850 1900 1950 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
number and productivity of active composers over previous 10 years
Figure 7: Number and productivity of active composers.
8
20%
40%
60%
80%
100% output percentiles of active opera composers over previous 10 years
0 1 2 3
9
2
4
6
8
10
12
mean median
Figure 9: Average time elapsed since active composers last premiered a work.
10
3 History of the Paris Opera
3.1 Management of the Paris Opera
The management of the Paris Opera varied considerably over its 350-year history. It can be roughly divided into four periods.
During the first period the State created a privilege but left its operation to others. The origin of the Paris opera lies in an exclusive privilege granted by the king in 1669 to Pierre Perrin, revoked in 1672 and granted to Jean-Baptiste Lully. The institution had an exclusive monopoly on French-language opera throughout the realm, which meant in practice that it charged licensing fees to any other theater that wished to perform sung music on a stage. Until 1749 the privilege was granted to a succession of entrepreneurs. In 1749 the privilege was ceded to the city of Paris, which either managed the theater directly or entrusted it to an entrepreneur. The government’s involvement consisted mostly in enacting general regulations, in particular two regulations of 1713 and 1714 governing the management of the opera.
From 1780 to 1831 came the second period, interrupted by the French Revolution, of direct management by the government. In 1780 the Opera was entrusted to the Master of Revels (intendant des Menus Plaisirs), an officer in the King’s Household. With the Revolution came the abolition of all privileges and the deregulation of theaters, and the Paris Opera lost its privilege. The Paris Opera continued as an institution. It was taken over by the city of Paris from 1790 to 1795 (during which brief period direct management, leasing, and management by the artists were all tried) but returned under government management, at first under the ministry of the Interior, then from 1803 under the Household of the sovereign (First Consul, Emperor, King). Under Napoleon the Opera regained the substance of its former privilege, namely the exclusivity over the operatic genre (1806) and payments from all other theaters in Paris (1811).
The third period begins in 1831, when the new regime of Louis-Philippe returned to the 18th century tradition of leasing to an entrepreneur, with a fixed subsidy. Private management was interrupted from 1854 to 1866, when serious financial difficulties led to an effective nationalization of the Opera. For a brief while the monarchical model of supervision by an officer of the Household was followed, but in 1864 the deregulation of theaters deprived again
11
the Opera of its legal privilege (payments from theaters had ceased in 1831), and the government returned to the regime of subsidized entrepreneurship.
The final period begins in 1939 when the government placed the two subsi- dized lyric theaters, the Opera and the Opéra-Comique under the Réunion des théâtres lyriques nationaux. The Opera has been a government-run institution ever since.
During the period of this study, the management of the opera oscillated between two regimes, depending largely on political and financial circumstances: direct supervision and leasing.
3.2 Direct supervision
In one regime, which only existed during monarchical periods, the government managed the theater directly, under the close supervision of the Minister of the Royal or Imperial Household. The responsibility for managing the theater was entrusted to either one person, a director, or to two individuals, a director and an administrator. Financial responsibility for profits and losses rested ultimately with the ministry. The government would edict regulations governing the operations of the theater. The minister or some lower-ranking official had direct control over many aspects of the theater, including choices of works to perform.
Préfet du Palais surintendant des spectacles (1 Nov 1807)
3.3 Leasing: the contracts
In the second regime, the theater was rented out to a private entrepreneur, chosen by the government, who took on all financial risks.3 The terms of his contract were called the cahier des charges.
The cahier des charges, terms of reference or, as I shall call them, the contracts, became a fundamental element of the leasing of the Paris Opera. The first contract was signed in 1831. The duration of contracts varied somewhat. the leases of 1831 and 1835 were for 6 years, the lease of 1841 was
3In 1874, after the theater burned down and the Opera moved to the Palais Garnier, the State agreed to allocated 2,400,000F to rebuild new sets for the new opera house, and the profits were split between the entrepreneur and the State; the losses remained the entrepreneur’s responsibility. This profit-sharing did not appear in any other lease.
12
for 8 years, that of 1847 for 10 years, those of 1866 and 1871 for 8 years, and from 1879 the norm became 7 years.
The contract specified the obligations of the entrepreneur. The basic terms were similar from contract to contract, with some variations and, in general, an increase in the obligations. From 1847 on, there was a general statement of the purpose of the Opera: the director was to manage it in a way that befitted its status as the premier lyric stage of France and to maintain it in “a condition of luxury that distinguishes it from all other houses from the point of view of the richness of sets and costumes as well as the number of artists” (1847). From 1866 it was not just the number, but also the talent of the artists; from 1871 it was to distinguish itself in “the selection” and from 1891 “the selection and variety of ancient and modern works,” The 1879 went as far as to state that “the Opera is not an experimental theater (théâtre d’essai) , it must be considered the museum of music,” although this formulation was not repeated afterward. In 1907 the directors were required “constantly to maintain in the repertoire the main works of classical composers.”
Until 1891 the genres were restricted to operas with orchestral recitatives (thus excluding spoken dialogue) with or without ballet, and ballets. From 1891 all kinds of lyrical and dance works were allowable, although an exception was made in the 1891 contract for the opéra comique (as defined in 1807) which remained the privilege of the theater of that name. From 1894 no genre restrictions were placed. The other meaningful restriction was on the number of new works to be performed. The requirements varied over time (see below). The quotas were yearly but (from 1835) evaluated on a 2-year period, and waivers could be obtained when the success of a particular work justified it. In principle a new work had to be staged every three months, and only one work by the same composer and librettist was allowed in a given year. Until 1891 the staging of all works required prior approval of the minister.
Other constraints were placed on the number of performances. A minimum of three performances per week was imposed, and until 1866 no more than four per week. The theater was to remain open continuously, unless the government decided that renovations of the theater were necessary. The total number of performances required increased over time, to reach 194 per year in the 1907 contract. In addition the director was allowed to organize concerts and balls.
Concern about the quality of performances translated into requirements on staffing. A minimum number of artists (singers, choir, dancers, ballet ensemble,
13
orchestra) was stated, and from 1841 to to 1871 minimum budgets for the choir, ballet ensemble and orchestra were specified. From 1879 this was replaced by minimum salaries in each category. Interestingly, however, although the engagement contracts for singers and dancers were standardized, no restrictions were placed on their salaries (in contrast with the regulations of Napoleon I, which placed upper limits on salaries).
The buildings (theater, workshops, and warehouses) and all materials were placed at the disposal of the entrepreneur. He was responsible for maintenance but not for major works or repairs, like any tenant. Inventories were drawn up at the beginning of the lease and the entrepreneur was obligated to return materials of equal value; any excess, in particular new sets and costumes, belonged to the State. Requirements were initially imposed to create new sets and costumes for all new works, but from the mid-1840s they were waived and eventually dropped. The director was required to insure the materials against fire, and a minimum insured value was specified from 1866.4
Until the 1850s disputes regarding the contract were adjudicated by the ministry’s commission on theaters, but in the 1850s recourse was granted to the Council of State and from the 1870s the contracts fell into the jurisdiction of regular courts.
The entrepreneur was required to deposit a bond. How he financed his operations varied: in the 1830s Duponchel was allowed to set up a limited partnership. This form of financing became more common from…
Abstract
Using a database of performances at five major opera houses from c1750 to c1950 I document that a shift in the performance of operas occurred in the mid-19th century, away from performing only recent works and for a short while, toward the emergence of a growing canon of known works and composers. The shift occurs roughly at the same time in all the theaters, suggesting that continent-wide factors were at play rather than national idiosyncracies.
Looking for economic forces to explain the shift, I document the financial history of one theater, the Paris Opéra, from 1803 to 1914. Like many other theaters at the time, it was mostly managed by for-profit entrepreneurs. Problems of rising costs and high risks were constant preoccupations. The main rise in costs appears to be in the wages of artists, particularly soloists, with increasing inequality and the emergence of a star system. A strategy of risk-avoidance may have driven directors toward programming works with known track record of success.
Apologies to EHES 2015 participants for this very rough, preliminary, and incomplete draft.
Keywords: (JEL ).
1 Introduction
This paper is a first venture into the economic history of opera. The first point is that a substantial shift in the performance of operas took
place in the middle of the 19th century. Before then, operas were produced in relatively large quantities and performed for a few years before being forgotten. There were a few exceptions, mostly local (Paris tended to have a culture of classics more than other opera houses; Vienna tended to preserve the memory of Mozart when he was forgotten everywhere else).
Can this be explained by economic factors? Of course, it is possible to ascribe this to preferences, or intrinsic quality. The latter explanation is simple: until 1850, no works worth keeping were produced. This is a little problematic: not only were good works produced before 1850, in fact a substantial part of the modern repertoire is made up of those works, rescued from oblivion over the years. Preferences might explain the fact, as follows: until the mid-1850, opera-goers valued novelty over quality. After 1850, tastes shifted toward quality. This explanation is only partial (why the shift?) and also does not explain why new works were constantly added to the repertoire after 1850.
My approach is to look for a documented shift in factors external to the opera industry. The intuition (yet to be modeled) is that increasingly high fixed costs made the performance of new works financially risky, and drove risk-averse entrepreneurs toward the safety of known successes.
To document this I turn to the Paris Opéra, not perhaps the most represen- tative institution but the one I could research relatively easily. From 1831 until the 1930s it was managed by for-profit entrepreneurs who leased the institution. Admittedly they operated under many constraints which I document. They also received a sizeable subsidy but, as I show, it was lump-sum and hence did not affect decisions at the margin. I reconstruct annual budgets from 1803 to 1914 and look at revenues and costs, as well as capacity and pricing. I focus particularly on salaries of artists, which was the main source of growth in costs until the 1880s (hence during the period of the shift I wish to explain).
1
Tentative speculations will be offered in lieu of a conclusion.
2 The Formation of the Operatic Canon
2.1 The Repertoires of European Cities
The starting point of the enquiry is a database of all operas performed in the major European opera houses:
• Paris: Opéra (1750–1945) • Paris: Opéra-Comique (1826–1914) • Naples: San Carlo (1738–1987) • Milan: La Scala (1778–1963) • London: King’s Theater – Covent Garden (1760–1957) • Vienna: Burgtheater / Kärntnerthortheater – Hofoper (1770–1944)
There are in fact two databases: one containing the identity and count of each opera performed in each year and each theater, and another database of operas with their characteristics: composer name and biographical information (birth and death date and place) and date and place of the first world performance. I then use the two databases to study the characteristics of the repertoire (in one sense of the word, namely, what is being performed in a given place at a given time).
Figure 1 displays the basic pattern. It plots the average age of the operas performed in each theater. The age of a work is the time elapsed since its first world performance, and the ages are weighted by the number of performances in each year. A very clear pattern emerges, with two regimes. Before the mid-19th century, the average age in most theaters is between 0 and 10. Paris is a bit of an outlier: one can discern periods of rising average age, or aging repertoire, one peaking in 1770 (shortly before the works of Gluck replaced those of Rameau), and another peaking in the 1820s. From the mid-19th century, the average age begins a steady rise everywhere, with a pause in the late 19th century when new works are being added, but then resumes and continues at a 45-degree angle through the 20th century.
This may seem surprising to those familiar with the modern repertoire of opera houses: there was a time when operas were performed as soon as they were composed, played for a few years, and then tossed out. Opera, even of
2
10
20
30
40
50
60
70
80
90
Comique La Scala London Paris San Carlo Wien
Figure 1: Average age of operas performed in major European opera houses.
the high-brow sort, was entertainment rather than culture, more like cinema today. Figure 1 dates the beginning of opera as canon to the mid-19th century.
The average is only one summary statistic. Many others can be computed. Figures 2 and 3 show the quantiles of the distribution. An important element of the pattern of aging becomes clear: new works are continuously being added, and some do join the repertoire, tending to lower the average age. But at the same time old works are also being resurrected and added to the repertoire, tending to counteract this lowering. Some vintages appear to contain more durable works, tracing parallel 45-degree lines in the age distribution.
Figure 4 shows the following statistic: in each theater and each 5-year period, I consider the works premiered, tally up their cumulative number of performances over time for those works, and show the average per cohort. a flat line means works from that half-decade are not performed, a rising line means works from that cohort are being performed. Some lines rise for a long time and then stop, indicating the emergence of “premature canons.”
Are the composers being performed dead or alive? Figure 5 shows the answer. The same broad pattern: there is a big difference between pre-1830 and post-1870, although again Paris is an outlier.
3
50
100
50
100
50
100
50
100
50
100
50
100
150 Wien
Figure 2: 25th, 50th, and 75th percentiles of the age of operas performed in major European opera houses. Age is on the vertical axis.
Although the pattern of aging is broadly similar across countries, there are important differences. One emerges when we consider the origin of works performed. Figure 6 shows the proportion of works performed in each theater that were premiered in that theater, that city, or that country. For this exercise contemporary rather than modern borders are used to defined countries, except for Italy, in which I included Lombardy (where Milan is located) and the Veneto (where Venice is located) even before their annexations in 1860 and 1866 respectively.
There is a pattern of internationalization of the repertoire through the 19th century, but it proceeds at different paces in the various cities, with London and Vienna being more open to foreign works earlier. Paris is the most closed city, with both the Paris Opera and the Opéra-Comique displaying a stronger home bias for longer.
4
50
100
50
100
50
100
50
100
50
100
50
100
150 Wien
Figure 3: 10-% quantiles of the age of operas performed in major European opera houses; age is on the vertical axis. The read line is the median.
2.2 Composer Productivity
A third database, which contains essentially all operas ever performed (some- where around 33,000 distinct works) allows me to point out another interesting pattern, on the supply side so to speak.
A big shift starts in 1800, and is completed by 1850: • the number of active composers doubles (“active” means he has already
premiered his first work, not yet his last) • composer productivity falls by half before the emergence of the canon • the time elapsed between premieres doubles in the same period • this is not sensitive to exclusion of lighter works or exclusion of “amateurs” (composers with only one work)
2.3 A hypothesis
As an economist, I look for economic explanations. I also share my tribe’s prejudice that preference-based explanations are, if not the last refuge of
5
50
100
50
100
50
100
150
200
200
400
600
50
100
100
200
300
400
Figure 4: Average cumulative performances by 5-year cohorts.
scoundrels, a deeply unsatisfying route. This doesn’t mean I won’t model demand, and accept the notion of a quantity-quality tradeoff. But I wish to keep preferences constant.
Nor am I interested in the “objective quality” view. As a lover of Monteverdi, Cavalli, Lully, Haendel, Rameau, Mozart, and many other pre-Verdi composers, I will not postulate that works produced after Il Trovatore were simply better.1
What I wish to explore is an exogenous shift in costs, which, combined with an undiversified, risk-averse entrepreneur, will bias programming choices toward works that are known to have a good track record, and avoid risky ventures on unproven works or composers.
The rest of this paper is a “deep dive” into the financial history of one institution, the Paris Opéra, over a century, a period that spans the shift that I wish to explain.2
1Sorry, Joel Mokyr. We’ll just have to disagree. 2Figures 7–9 suggest that something interesting is also happening upstream, on the production side.
6
1760 1780 1800 1820 1840 1860 1880 1900 1920 1940 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Comique La Scala London Paris San Carlo Wien
Figure 5: Share of dead composers being performed.
7
0.6
0.8
0.2
0.4
0.6
0.8
0.2
0.4
0.6
0.2
0.4
0.6
0.8
0.2
0.4
0.6
0.8
0.2
0.4
0.6
0.8 Wien
Figure 6: Share of works locally premiered, by theater, city, and country.
1650 1700 1750 1800 1850 1900 1950 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
number and productivity of active composers over previous 10 years
Figure 7: Number and productivity of active composers.
8
20%
40%
60%
80%
100% output percentiles of active opera composers over previous 10 years
0 1 2 3
9
2
4
6
8
10
12
mean median
Figure 9: Average time elapsed since active composers last premiered a work.
10
3 History of the Paris Opera
3.1 Management of the Paris Opera
The management of the Paris Opera varied considerably over its 350-year history. It can be roughly divided into four periods.
During the first period the State created a privilege but left its operation to others. The origin of the Paris opera lies in an exclusive privilege granted by the king in 1669 to Pierre Perrin, revoked in 1672 and granted to Jean-Baptiste Lully. The institution had an exclusive monopoly on French-language opera throughout the realm, which meant in practice that it charged licensing fees to any other theater that wished to perform sung music on a stage. Until 1749 the privilege was granted to a succession of entrepreneurs. In 1749 the privilege was ceded to the city of Paris, which either managed the theater directly or entrusted it to an entrepreneur. The government’s involvement consisted mostly in enacting general regulations, in particular two regulations of 1713 and 1714 governing the management of the opera.
From 1780 to 1831 came the second period, interrupted by the French Revolution, of direct management by the government. In 1780 the Opera was entrusted to the Master of Revels (intendant des Menus Plaisirs), an officer in the King’s Household. With the Revolution came the abolition of all privileges and the deregulation of theaters, and the Paris Opera lost its privilege. The Paris Opera continued as an institution. It was taken over by the city of Paris from 1790 to 1795 (during which brief period direct management, leasing, and management by the artists were all tried) but returned under government management, at first under the ministry of the Interior, then from 1803 under the Household of the sovereign (First Consul, Emperor, King). Under Napoleon the Opera regained the substance of its former privilege, namely the exclusivity over the operatic genre (1806) and payments from all other theaters in Paris (1811).
The third period begins in 1831, when the new regime of Louis-Philippe returned to the 18th century tradition of leasing to an entrepreneur, with a fixed subsidy. Private management was interrupted from 1854 to 1866, when serious financial difficulties led to an effective nationalization of the Opera. For a brief while the monarchical model of supervision by an officer of the Household was followed, but in 1864 the deregulation of theaters deprived again
11
the Opera of its legal privilege (payments from theaters had ceased in 1831), and the government returned to the regime of subsidized entrepreneurship.
The final period begins in 1939 when the government placed the two subsi- dized lyric theaters, the Opera and the Opéra-Comique under the Réunion des théâtres lyriques nationaux. The Opera has been a government-run institution ever since.
During the period of this study, the management of the opera oscillated between two regimes, depending largely on political and financial circumstances: direct supervision and leasing.
3.2 Direct supervision
In one regime, which only existed during monarchical periods, the government managed the theater directly, under the close supervision of the Minister of the Royal or Imperial Household. The responsibility for managing the theater was entrusted to either one person, a director, or to two individuals, a director and an administrator. Financial responsibility for profits and losses rested ultimately with the ministry. The government would edict regulations governing the operations of the theater. The minister or some lower-ranking official had direct control over many aspects of the theater, including choices of works to perform.
Préfet du Palais surintendant des spectacles (1 Nov 1807)
3.3 Leasing: the contracts
In the second regime, the theater was rented out to a private entrepreneur, chosen by the government, who took on all financial risks.3 The terms of his contract were called the cahier des charges.
The cahier des charges, terms of reference or, as I shall call them, the contracts, became a fundamental element of the leasing of the Paris Opera. The first contract was signed in 1831. The duration of contracts varied somewhat. the leases of 1831 and 1835 were for 6 years, the lease of 1841 was
3In 1874, after the theater burned down and the Opera moved to the Palais Garnier, the State agreed to allocated 2,400,000F to rebuild new sets for the new opera house, and the profits were split between the entrepreneur and the State; the losses remained the entrepreneur’s responsibility. This profit-sharing did not appear in any other lease.
12
for 8 years, that of 1847 for 10 years, those of 1866 and 1871 for 8 years, and from 1879 the norm became 7 years.
The contract specified the obligations of the entrepreneur. The basic terms were similar from contract to contract, with some variations and, in general, an increase in the obligations. From 1847 on, there was a general statement of the purpose of the Opera: the director was to manage it in a way that befitted its status as the premier lyric stage of France and to maintain it in “a condition of luxury that distinguishes it from all other houses from the point of view of the richness of sets and costumes as well as the number of artists” (1847). From 1866 it was not just the number, but also the talent of the artists; from 1871 it was to distinguish itself in “the selection” and from 1891 “the selection and variety of ancient and modern works,” The 1879 went as far as to state that “the Opera is not an experimental theater (théâtre d’essai) , it must be considered the museum of music,” although this formulation was not repeated afterward. In 1907 the directors were required “constantly to maintain in the repertoire the main works of classical composers.”
Until 1891 the genres were restricted to operas with orchestral recitatives (thus excluding spoken dialogue) with or without ballet, and ballets. From 1891 all kinds of lyrical and dance works were allowable, although an exception was made in the 1891 contract for the opéra comique (as defined in 1807) which remained the privilege of the theater of that name. From 1894 no genre restrictions were placed. The other meaningful restriction was on the number of new works to be performed. The requirements varied over time (see below). The quotas were yearly but (from 1835) evaluated on a 2-year period, and waivers could be obtained when the success of a particular work justified it. In principle a new work had to be staged every three months, and only one work by the same composer and librettist was allowed in a given year. Until 1891 the staging of all works required prior approval of the minister.
Other constraints were placed on the number of performances. A minimum of three performances per week was imposed, and until 1866 no more than four per week. The theater was to remain open continuously, unless the government decided that renovations of the theater were necessary. The total number of performances required increased over time, to reach 194 per year in the 1907 contract. In addition the director was allowed to organize concerts and balls.
Concern about the quality of performances translated into requirements on staffing. A minimum number of artists (singers, choir, dancers, ballet ensemble,
13
orchestra) was stated, and from 1841 to to 1871 minimum budgets for the choir, ballet ensemble and orchestra were specified. From 1879 this was replaced by minimum salaries in each category. Interestingly, however, although the engagement contracts for singers and dancers were standardized, no restrictions were placed on their salaries (in contrast with the regulations of Napoleon I, which placed upper limits on salaries).
The buildings (theater, workshops, and warehouses) and all materials were placed at the disposal of the entrepreneur. He was responsible for maintenance but not for major works or repairs, like any tenant. Inventories were drawn up at the beginning of the lease and the entrepreneur was obligated to return materials of equal value; any excess, in particular new sets and costumes, belonged to the State. Requirements were initially imposed to create new sets and costumes for all new works, but from the mid-1840s they were waived and eventually dropped. The director was required to insure the materials against fire, and a minimum insured value was specified from 1866.4
Until the 1850s disputes regarding the contract were adjudicated by the ministry’s commission on theaters, but in the 1850s recourse was granted to the Council of State and from the 1870s the contracts fell into the jurisdiction of regular courts.
The entrepreneur was required to deposit a bond. How he financed his operations varied: in the 1830s Duponchel was allowed to set up a limited partnership. This form of financing became more common from…
Related Documents
