Introduction to the Berliner Section of James R. Smart's Sousa Band Discography.
James R. Smart's introduction to the Berliner Gramophone section of his Sousa Band discography provides a detailed and insightful look at both the recordings of the band for Berliner and the Berliner Company's two disc numbering systems.
The Berliner Gramophone Company
From: James Robert Smart. The Sousa Band, A Discography. Washington: Library of Congress [For sale by the Supt. of Docs., U.S. Govt. Print. Off.], 1970.
The first disc recordings of the Sousa Band were made for the first manufacturer in this field, the Berliner Gramophone Company of Philadelphia and its founder, Emile Berliner. In view of the great importance of the Berliner company's work in connection with the evolution of the disc record industry, it is regrettable that the recording logs and other details of the business have not been found. Therefore, in order to throw as much light as possible on the Sousa Band's work for Berliner, it is necessary to rely upon what few facts can be established and then to relate the band's recordings to them. Fortunately, it was often the practice of the company to inscribe the recording date into the etched or embossed label area of the disc. Less often, it also indicated the location at which the recording was made—either Washington, Philadelphia, or New York—and also at times the name of the conductor. Although only a small percentage of the Sousa Band's records have been located and this information extracted from them, these few make it possible to place the origin of all the known records within a reasonable time span.
At first glance, it would seem that the system of catalog numbers used by Berliner constitutes an important aid for establishing the period during which a certain record was made, and so it does for later releases. But for records made before 1899 this method must be used with some caution. First, it must be pointed out that Berliner began numbering his discs starting with the earliest commercial releases in Washington in 1893. This must be considered an innovation, since none of the cylinder manufacturers of the same period were placing permanent numbers on their productions.
When the Philadelphia factory and studio were opened, the numbering system was continued, but evidence shows that some titles were withdrawn about this time and their catalog numbers used for different works. This numerical system evidently began with record 1 and progressed to 999. Then come large groups of records with numbers ranging from 1200 to (at least) 1980. So far as the many gaps in our knowledge permit us to judge, all of these records appear to have been issued before the fall of 1898. During this same' period, Berliner also issued records using the block numbering system, such as 2500 for piano solos; 2900 for Swedish language recordings; 3300 for trombone solos with piano accompaniment; and several others.
It will be remarked that there is a gap in the catalog numbers between 1000 and 1199. These numbers may have been pre-assigned for the use of William Barry Owen and Frederick W. Gaisberg in establishing a branch of the Berliner concern in England, and the first English made Berliner records, released in August 1898, do, in fact, have numbers beginning with 1000 and continuing up through 1194. Beginning with 1200 both Berliner branches go their separate ways.
Some time in late 1898, or possibly at the beginning of 1899, Berliner began a completely new numbering system, this time commencing with record 01 and proceeding up to at least 01304. This last numbered record was made around the middle of 1900. Since the company suspended operations under a court injunction issued on June 25, 1900, the record numbers probably do not go much higher. During this period, too, Berliner continued to issue some records in the block numbering system, and the entire 8000 series of band recordings seems to have originated after the 01 system was begun. Taking both numerical systems and adding to them the records issued in the block system, one finds that Berliner must have issued domestically something more than 3,000 records.
The above quantity is misleading, however, and the actual number of records issued by Berliner is probably considerably more. The reason for this is that the company seems to have released a large number of remakes or re-recordings of a title made either by the same artist or a different one, but continuing to use the same catalog number. There are even some cases where three different artists recorded the same title and all the records were released with the same catalog number. Some of the remakes involve Sousa Band records. A number of titles were made not only by it but also by the Banda Rossa and some other bands of the day. In addition; the Sousa Band remade some of its own earlier recordings. For instance, Meeting of the Blue and the Gray was recorded around August 1897 and issued as record 128. It was re-recorded in September 1898 and again issued as record 128. After the beginning of the 01 numerical system, it seems to have been customary to give remakes new catalog numbers, although the older method continued to be used to some extent.1
To maintain control over the remakes, Berliner placed an identifying symbol after the catalog number consisting of the letters V through Z, and sometimes ZZ. It is not at all clear just how this identification system works, and it is one of the most mystifying aspects of Berliner's business.2 The letters are usually not printed in the company's catalogs and can only be found on the record itself.
Another complication connected with the catalog numbering systems is the fact that of the first 149 records issued in the earlier system, the titles and performers of about 90 of them are known. They are all band recordings. This could be happenstance, or it could be the result of design. If we examine the titles and names of performers for all known records carrying numbers between 1 and 1000, a rather definite grouping pattern begins to emerge. The earliest Berliner records seem to have been issued in groups as follows: 1-149, band recordings; 150-199, popular songs; 200-374, various instrumental solos, duets, etc.; 375-449, miscellaneous traditional songs, comic songs, whistling recordings; 450-499, banjo solos and duets; 500-699, popular songs; 700-709, fife, drum and bugle corps; 710-749, popular songs; 750-819, unidentified; 820-849, brass quartets; 850-899, vocal quartets; 900-999, opera arias, traditional songs, and popular songs. Following the gap from 1000 to 1200 one does not find such grouping by type of performing medium or type of composition, and the emphasis now seems to be upon individual performers. For instance, records 1450-1499 are all by the Metropolitan Orchestra, In like manner, the records released in the block numbering system are also oriented toward the individual performer.
Releases in the 01 numerical series, similar to the later releases as outlined above, do not indicate any grouping, except by individual performer. The series of Sousa Band recordings released on records 0175 through 0231, and again from 01169 through 01208, apparently reflect the current custom of releasing in a unified body all recordings resulting from one recording session or series of sessions. The records from 01169 through 01208 were all made at a series of recording sessions during the second week of April 1900. In these cases, the individual catalog number almost surely functions as a matrix number and as such indicates the approximate order of recording.
If the numbering systems used by Berliner seem unnecessarily complicated, it should be remembered that before the turn of the century the record industry was still in its infancy. Berliner seems to have been first to apply numbers to his records, and when he began in 1893 he could not have had any idea how far he would be able to go in the business, or how many records he would produce. Also, the idea of rejectIc.ing a performance, or recording a work two or three times at a session and then issuing only the "take" that seems to be the best, was probably foreign to his mind, as it was to all the other record manufacturers of his day.
It should now be obvious that the catalog numbers of Berliner records made before 1899 cannot be used as a means of arranging the records in chronological order; one should not be surprised to learn that a copy of 979 bears the date April 15, 1896, while a copy of 259 carries the date October 9, 1897.
From information now available, it appears that the Sousa Band recorded for Berliner at these times and locations:
August-September 1897, probably in New York, since at that time the band was playing its summer engagement at Manhattan Beach.
Early April 1898 in New York. At that time the band was playing Sunday evening concerts at the Metropolitan Opera House. In mid-April it began a lengthy concert tour.
Early September 1898, probably in New York. The band left for the Pittsburgh Exposition on September 6.
April 22, 1899, in Philadelphia. On tour at that time, the band played concerts in Philadelphia on April 21 and 22.
June 6 through 9, 1899, in New York. The band opened at Manhattan Beach in the middle of the month.
April 9 through 13, 1900, in Philadelphia. The band finished a tour on April 7. On April 22 it played a farewell concert in New York and on April 25 it sailed for Europe.
With regard to the final recording session, a Berliner catalog issued some time after April 1900, and possibly the company's last full catalog, states:
We present a new list of records by Sousa's Famous Band, made just before its departure for Europe to take part in the Paris Exposition as the Official Band of the United States Government.
If this statement is to be taken literally, then many remakes are indicated here because a large number of the titles in the catalog were made by the band as early as 1897. It would also indicate that the band recorded a total of 84 titles during the five days. This would by no means be impossibility. The 7" Berliner disc had a playing time of about two minutes. However, only one of the records with a low catalog number has turned up bearing a 1900 recording date, so that it seems that, while some records were undoubtedly remade, the last recording session consisted primarily of new recordings, those issued with catalog numbers 01169 and higher.
The same 1900 catalog states that the band is playing "under the personal direction of Mr. Arthur Pryor, the great trombone soloist and assistant conductor to Mr. Sousa." The records issued at that time bore Pryor's name as conductor too. Earlier catalogs say nothing about the conductor, only stating that the recordings were made "by special permission from John Philip Sousa." The records themselves indicate that Henry Higgins, as well as Pryor, conducted for Berliner recordings. The earliest releases, those dating from 1897 and 1898 and issued with catalog numbers below 149, and also those issued in the 8000 block, seem to have been conducted by Higgins, although remakes of some of these were conducted by Pryor; while records dating from the years 1899 and 1900 and name Pryor as conductor.
1. It is interesting to note that no cases of remakes have been found involving records issued in the block numbering system.
2. Letter suffixes are also used to identify different sections of a work. Record 901X, for instance, is the first verse of Le Marseillaise, while record 90lY is the second verse.