Part of my 1861 Toronto census project has involved matching people in the census to entries in city directories of the time, namely that of Caverhill in 1859-60 and of Mitchell in 1863-64. This has proved very interesting. There is evidence of a lot of movement around the city, as well as of people coming into the city and disappearing from it. From the census we learn a person’s age and his birthplace. From the directories we learn the kind of business he may be running or employed in. We may discover that a woman has been widowed during the five-year period. And I find hints that I may have misinterpreted a surname when transcribing the census. It has certainly been a worthwhile exercise.
In order to use the directories efficiently I have had to digitize them. Digitizing is a word we have heard in connection with computers over the past ten years, but do we know what it means? Taken simply, it is the process of copy-typing the directory entries onto a computer, and, at the same time, breaking down each line into its component parts of surname, given name, occupation and address. If you have transcribed a census or a list of burials in a Family Search project, you have digitized the manuscript information on the original forms.
Why should we have to do this for printed books? Can’t books be scanned and read by optical character recognition? Well, many more recently printed ones can, but 19th-century type does not lend itself to the procedure. The letters are just not sharp enough for a camera to identify each as one of the twenty-six used in our English alphabet. I have tried OCR programs on pages of old data and come very close to turning the air blue trying to make sense of the software’s interpretation. It might take a long time, but old fashioned copy-typing is the better plan of attack.
Since I was transcribing the directories myself I was able my own choices for headings. My initial reason for looking at the directories was that some enumerators had omitted addresses from the census forms. I wanted to know where these mystery areas actually were. For my purposes it was best to separate the directory addresses into streets and numbers and add in the city ward where the streets were located. From this I could make up a list of inhabitants, street by street, and house by house.
Both directories had a number of foibles. Caverhill’s inability to keep his records in strict alphabetical order was frustrating. I am suspicious that a pile of entries at the end of the alphabet went astray before the directory was printed. Initial letters W and Y are very poorly represented. He also had a bad habit of omitting the “East” or the “West” from streets that crossed Yonge. An address like 394 Queen could either be around Denison Avenue or between Sumach and River Streets, a very long way apart.
Mitchell’s Directory of 1864 had two parts and in one the households were arranged by streets. I started my database using this, but then found it only contained house owners or the heads of the tenant families. His alphabetical one contained not just heads but other people in the house: tenants, boarders, mothers-in-law, sons (and daughters) who were working or who were students in law or at university. Sometimes the addresses in the street directory differed from those in the alphabetical directory indicating that the family had moved house sometime during the compilation process. The alphabetical listing usually contained the later address. Mitchell had obtained staff lists provided by large employers. Unfortunately, many people’s addresses were missing. But the directory had an extra advantage: the entries for most widowed women contained the phrase “widow of John” or “widow of William”. How useful to the genealogist.
We ought not to complain about the imperfections in a mid 19th century city directory. Picture how it must have been developed—without electricity, typewriters, and all the other paraphernalia that we have depended on for the last hundred years. Until the directory was finally typeset, all the listings must have been done with pen and paper. It would have been easy to misread a colleague’s handwriting or even one’s own on a cloudy day in a dusty office. Omitting a letter or a cross on a “t” can make a lot of difference. Were entries made on foolscap pieces of paper or on slips that fitted in a shoebox? (Given the number of shoemakers in Toronto at the time, they must have had shoeboxes.) I hope it was the latter. Corrections are much easier to make that way. However, the typesetters might have insisted on the listings being on sheets of papers—and the rewriting would be another step in the procedure where errors could creep in.
Today I have three “digitized” databases illustrating Toronto’s population over a five-year period at the beginning of the 1860s. I hope it will prove to be a more useful series of documents than any one of the three taken individually.