Contract Date Versus Completion Date

The contract date is the day the builder and the buyer signed the agreement to build, and it is the date a published opus list can carry in its year column. The archived Ernest M. Skinner & Son and Aeolian-Skinner opus lists say so in their own notes: the years shown indicate when contracts were signed and do not reflect years of installation. That note explains apparent inconsistencies between a list year and the year an instrument was dedicated. Where a source gives only the year, call it a contract year rather than a date.

Contract dates survive well because a contract is a legal and financial document, and the record series built on contracts are dated by them. The finding aid for the M.P. Möller collection at the Organ Historical Society’s Library and Archives lists contract ledgers indexed by opus number and covering a contract-period year range. A Schantz instrument recorded in the Pipe Organ Database as opus number PW266 draws its 1918 date from a 1918 contract held in the Schantz company archive. That shows the archive can yield a number and a contract date for at least one early instrument, not that it can do so for every instrument on request.

Construction and shipping dates mark work in the factory and the departure of the finished parts, and they fall between the contract and the installation. The Möller collection keeps them apart from the contracts: its sales and shipping ledgers record the later commercial and dispatch events, so the same opus number can legitimately be dated to two different years depending on which series is cited. The shipping ledgers run from 1928 to 1979 in date order rather than opus order, so a reader who knows only the opus number cannot go straight to a shipping entry.

A construction-completion date is a related but distinct point. Létourneau’s Opus 134 for the Michigan State University Alumni Chapel had its construction completed in December 2020, about sixteen months before the instrument was dedicated. Whether a recorded date is a contract, shipping, installation or completion date is the first thing to settle before a run of dated instruments can be lined up against anything else, such as a change in a builder’s action technology; a series of bare years cannot support that kind of comparison.

A completion date records the point at which the instrument existed as a finished organ, and an installation date records the event of erecting it in the building; use the second term only where the record names the installation itself. Both come after the contract, and the gap can be short or long. Reuter Opus 227 in Portland, Oregon, is documented by a commercial sample-set publisher, working from Organ Historical Society data and the notes of Michael Barnes, with a contract date of 28 December 1926, a scheduled delivery of September 1927, and probable installation not before 1928: about a year and a quarter between contract and likely installation on one instrument.

Johnson Opus 100 shows a shorter interval. The parish history of St. Peter’s Episcopal Church in Hebron, Connecticut, records a contract dated 30 January 1860 and installation in May 1860, and says the church holds the original contract, a specification sketch and correspondence. That account quotes documents the parish says it holds; they have not been examined independently and their present custody is not verified, so the four-month interval rests on the parish’s own reading of its papers.

Completion dates are also what an institutional application can ask for. The Organ Historical Society’s Historic Pipe Organ Awards application treats the opus number as an optional field recorded only if known, and accepts a completion date given as a year alone. A year entered there is a completion year, a different quantity from the contract year on a builder’s list, and the two should not be copied into one column as if they were the same.

A dedication date records the public service or recital that marked the instrument’s acceptance, and it is neither the contract date nor the completion date. Létourneau Opus 134 at the Michigan State University Alumni Chapel was dedicated on Sunday, April 10, 2022, in two recitals, documented by the university’s own dedication-recital program and a separate university news article; its construction had been completed in December 2020, roughly sixteen months earlier.

A dedication date is one date an institution may print. A 2022 survey of the Casavant firm by Michal Szostak in The Organ describes the 1891 dedication of opus 26 at Notre-Dame Basilica in Montreal as the firm’s international breakthrough, dating the event by its dedication rather than by any contract. The trade press treats completion or dedication as its own field, separate from the contract: each entry in The Diapason’s New Organs column records the building firm and its town, the opus or job number, the venue and city, the completion or dedication date, the stoplist by division, the console specification and action type, and a builder’s narrative, all supplied by the building firm. A rededication date marks the same kind of event held after later work, and it belongs with the later-work dates below.

Later-work dates record rebuilding, restoration or additions carried out after the original installation, and they date the work rather than the organ. A builder’s rebuild list may give such a project a single year even when the work ran in more than one campaign, and the Létourneau rebuilds and restorations list publishes no note on what its Restored in year denotes, whether completion, contract, invoice or something else. One case where that single year disagrees with the dated trade-press account is set out under the disagreeing-records section below.

Later work also leaves dates on the instrument itself. A replaced console can carry a date later than the organ, which is the subject of Why the Date on a Console May Not Be the Date of the Organ, and a rebuilding firm can add its own plate and number, which is covered in Rebuild Plates, Secondary Opus Numbers, and Layered Instrument Identities. Here those dates count only as later-work dates; the two mechanisms behind them are explained on the pages just named.

A contract date establishes that an agreement to build existed on that day, with the specification and price it contained, and it fixes the instrument’s place in the builder’s numbered sequence, since the contract ledgers and published lists that carry those numbers are dated by that event. It supports a statement of the form contracted in 1921. It does not establish when the instrument was finished, shipped, installed or first heard, because each of those is a later event with its own record; Reuter Opus 227, contracted in December 1926 and probably not installed before 1928, would carry the earlier year on a contract-dated list.

Where a list says its years are contract years, as the archived Ernest M. Skinner & Son and Aeolian-Skinner lists do, the year cannot be read as a completion or installation year, and a parish history giving a later dedication year is not contradicting the list. How the remaining columns of such a list are read, and what an entry does and does not say about the instrument, is the subject of How to Read an Organbuilder Opus List.

Any statement about when the instrument was playable, when it arrived in the building, or which year an anniversary should mark needs a completion, installation or dedication record, not a contract. The same applies to comparisons across instruments: placing a builder’s numbered output against a change in its technology requires, as a first condition, that each instrument carry a date whose meaning is stated, whether contract, shipping, installation or completion.

Where to find those records follows from the record series. A trade-press New Organs column records a completion or dedication date supplied by the builder; an institution’s own dedication program or news release dates the dedication; a parish history, or a contract and correspondence kept in parish custody, may supply the installation month; and an awards application asks for a completion year. The Organ Historical Society’s Pipe Organ Database does not publish what its year field means, so a database year cannot be assumed to be an installation year.

These cases establish the method and its limits. They do not prove a conclusion about any particular instrument, which needs that instrument’s own records and, where the question turns on fabric, its physical evidence.

A one-year difference between a published contract year and a database year is the most common kind of divergence, and it can reflect two different events rather than an error. Matching the contract years on the archived Skinner Organ Company opus list against the years in the Organ Historical Society’s Pipe Organ Database for the same opus numbers, as of September 2026, gives 590 matched records: 542 agree exactly, 25 show the database year one year later, 4 show it one year earlier, and the remaining 19 diverge by between two and twenty-six years.

That comparison is a derived figure with limits. The database does not state what its year field means, and records flagged as relocated or rebuilt, or carrying a different original builder, were excluded. Read it as an indication that divergence is usually zero or one year and rarely more than two, not as a measured installation lag, and not as an error rate for any other list or database.

Named cases run in both directions. Skinner Opus 318, Washington Street Methodist Church, Columbia, carries contract year 1920 and database year 1922; Opus 347, St. Mark’s Episcopal Church, New Britain, carries contract year 1921 and database year 1922. Opus 331, Eastman School of Music, Rochester, carries contract year 1921 and database year 1920, and a database year that precedes the contract year cannot be explained as time between contract and installation; for that instrument at least one of the two records is wrong. Divergence is therefore a discrepancy to adjudicate from the instrument’s own records, not automatic evidence of a lag. The Möller ledgers show the same thing from the archive side: the same opus number can legitimately be dated to two different years depending on whether the contract ledger or the shipping ledger is cited.

A single year attached to a restoration or rebuild can stand for a project that ran in phases over several years, so it can disagree with a dated account without either record being false. As of September 2026, the Létourneau rebuilds and restorations list gives the Séminaire St-Joseph project in Trois-Rivières the single year 2017. The Diapason’s news item of 23 September 2016 describes the same project as two campaigns: a wind-system restoration in 2014, and a comprehensive windchest, console and reed restoration begun in July 2016 and expected to finish that fall. The list publishes no note on what its Restored in year denotes, so the difference cannot be attributed to any stated convention; both statements stand as their sources give them, and neither is averaged into the other.

The same list dates the Winthrop University project on Aeolian-Skinner Opus 1257 to 2009 alone, where a trade-press feature in The Diapason, co-authored by a member of the firm that did the work, gives 2007 to 2009. One project of each kind shows that a list year can compress phases and can post-date the trade-press account; it does not show how often that happens. Whether the later work also gave the instrument a second plate or number is a separate question, answered by the rebuild-plates page named above.

Record every date with its type and its source, so that a later reader can see which event the year stands for. Write contract year 1921, from the Skinner Organ Company opus list, or dedicated April 10, 2022, from the university’s dedication program, rather than a bare year; where a source gives only a year with no stated meaning, say that too. Keep two disagreeing years as two entries with their own attributions rather than choosing one or splitting the difference. A discrepancy that is written down can be resolved when the instrument’s own contract, shipping record or dedication program is found; an averaged year cannot be checked against anything.

This is the discipline the primary record series already follow: contract ledgers, shipping ledgers, trade-press columns and awards applications each name the event they date, and a reader’s notes should be no less specific. When the dating question moves from the records to the fabric, to a console that carries a later date, or to an instrument that has been given a second number by a rebuilder, the answer lies with the two pages named above and with physical inspection rather than with the record alone. Pipe-Organ Opus Numbers and Date Evidence sets out where each of those questions is answered.