Historians treasure their sources, and Montréal is particularly rich in raw materials for exploring the urban past—all the way back to the 1650s! The MAP project focuses on the half-century from 1848 to 1912. The name refers not only to the wealth of maps available, but also to “Montréal, l'avenir du passé” (Montréal: The Future of the Past), the idea that the past is part of our future and that our future needs to take the past into account. The site therefore offers a chance to explore our sources and the information they provide on the city between 1848 and 1912.
Here, we provide an initial overview of our sources and databases, along with examples of each type. What kinds of maps and data are available? Remember, we see maps as a kind of playground, and by switching among four kinds of maps and four kinds of databases, you can race from the swing to the slide to the merry-go-round.
Late nineteenth and early twentieth-century maps are great sources of information. Here, we feature the atlases published by C.E. Goad in 1881 and 1912. They are rare and fragile and bulky, but thanks to modern imaging techniques, you can appreciate all their fine details. In fact, the maps are so detailed they show whether houses had passageways for horses and carts, and whether they were made with wood or a more fire-resistant material—stone, brick, or “brick cladding” (cased in brick).
Using the source images, we have created maps that not only convey the available information but also make the data more readable and manageable. These “map layers” allow you to zoom in on individual city blocks. Or you can patch all the plates together and zoom out for a bird’s eye view over the river and across the railway yards.
You can also overlay two maps to discover relationships. For instance, you can display images from Goad’s 1912 Atlas beneath a map of today’s city. Adjust the zoom according to your needs using the in (Zoom in and out ) buttons. (Goad’s maps were originally published with a scale of 1 inch to 400 feet.)
Our maps can also combine data from multiple sources. For example, much of the information on Belonging, Density and Domestic Service presented in the Home Base section comes from the censuses of 1881 and 1901. However, pinning the data to specific locations and properties required information from tax rolls.
Thematic maps let you analyze themes by grouping several sets of data. For instance, how closely did the pattern of rents match the distribution of people by language or religion? Are you interested in a single household, a neighbourhood, or an entire ward? The maps can also show how patterns changed over time. How much of the city’s land was built up in 1901, compared to 1881? How did newcomers, such as Jewish immigrants who arrived in the 1890s, fit into demographic patterns?
Each source has its own logic. How was it created? For whom? For what purposes? This logic imposes constraints on what you can do with a particular source, while also offering insight into other sources.The tables we produced using the 1881 and 1901 censuses are divided into two parts: households and household members.
In five out of six households, everyone belonged to the same “nuclear family”—a father, a mother, and their children (although such households sometimes included children from a previous marriage, who had a different family name). One household in ten had a servant, and one family in six was “extended” to include grandparents, in-laws or boarders. There were several dozen “specialized” households (the presbytery, the bishop’s palace, the WCTU’s little shelter for women, etc.) and some two dozen “oversized” households (the seminary, a boarding school, the Windsor Hotel, several hospitals, the orphanage, the jail, the convents). By 1901, nearly 4% of the population lived in institutional households. However, some of the census worksheets are hard to interpret. Which residents were “housed” in an institution? Which ones were servants and other staff? Which ones were supervisors and professionals?
The 1901 Census we have a special problem. Although all households appear in an index (most of them with an address), the index provides very limited information on household members (name, age, birthdate, and marital status). Additional information is available on microfilm or online. And thanks to our many collaborators, we have digitally fleshed out the details for individuals in a quarter of the city’s households. So as soon you start looking at census variables like religion, language, and occupation (the interesting ones), you have to work with samples. These samples were created by six different research teams that collected data for different purposes and using different sampling strategies. The Home Base section uses information supplied by these teams to provide information on all residents of 33 small neighbourhoods. For more details on sampling strategies and the limitations of our 1901 census databases, make your way to The Kitchen.
Tax rolls provide two kinds of information, starting with data on individual properties: the lot size, the value that was subject to property tax, and the name(s) of the owner(s) who received the tax bill. It is fairly easy to distinguish between corporate owners and private individuals, all of whom were classified by religious affiliation: Catholic, Protestant, or “neutral”. This curious practice was necessary to channel school taxes to the Catholic and Protestant school boards. MAP has created digital versions of the tax rolls for 1848, 1860, 1880, and 1903.
Tax rolls also provide information on a property’s occupants, whether they were the owners, a tenant family, or a business tenant who used it as a shop or work space. Occupants paid a tax based on the floor area they occupied, as recorded in a “mini-census” conducted each June. This document did not record the number of family members, employees, or boarders who lived at a property. But since occupants who kept horses—tavern keepers (who ran inns, hotels, and taverns), carters, and cabdrivers—were charged for a special licence, the tax rolls do identify them reliably. MAP has created digital versions of the occupant rolls for 1848, 1860, and 1880 (but not 1903).
Tax rolls are key to the entire MAP project because of the precision with which they catalogue all the properties in the city. The 1880 tax roll lists 12,000 lots, each with its cadastral number and street addresses. They were recorded sequentially, for the convenience of tax assessors and collectors.
Lovell's directories of Montréal businesses are a wonderful source of additional details on companies and professionals. The originals are available online from the BANQ. In the 1881 and 1901 editions, names are listed in alphabetical order or by street and house number. MAP has created digital versions for 1848 and 1881. You can search by occupation and for women householders, corporations, and partnerships. And because entries are linked to the map, you can limit searches to a specific city block or street segment. In 1891, John Lovell even published his own “census” that paid close attention to religious affiliations.
A Mixed Bag: The power of the MAP project resides in the links that make it possible to identify the same person across multiple sources, or to use information from one database to interpret another. But the moment you start making these links, you start encountering problems. Some people just can’t be linked; they go missing in one source or another. Sometimes, we just get it wrong, and users like you are sure to discover some of our mistakes.
The 1901 Census provides addresses, but they appear on a separate page (schedule) of the census manuscript. Some of these addresses are ambiguous. Some are missing the house number, and there are others that we could not find on any map. In some cases, we assumed that multiple families lived in the same house… and we may have been wrong. In any case, we still have 1000 households that do not seem to be located in the right census tract.
The 1881 Census provides no addresses. However, we have been able to locate between 85 and 90% of all households by linking names to properties listed in the tax roll (or sometimes in Lovell’s Directory). This forces us to ask the question: How are the households we cannot place different from the ones on the map? We doubt there is a serious selection bias, but we cannot be entirely sure.
| For additional information or to make suggestions,
Dept. of Geography
805 Sherbrooke St. W.
Montréal, QC, H3A 0B9
Robert C.H. Sweeny
Dept. of History
St John's TN A1C 5S7