An extremely common feature of golden age detective mysteries is the presence of servants in a household. They acted as witnesses for the police, to place people at the scene of a crime as well as to provide alibis. They were also invaluable sources of information when discretely pumped. It is very difficult to come up with any modern equivalent, though, at least outside of exceedingly rich households.

In real life, servants occupied a curious niche in British culture during the early 1900s; with the rise of the middle class servants were relatively commonplace, since the middle class was comparatively wealthy and the transition from farms to a modern economy was still underway, supplying a large number of people who had few specialized skills but just as much need to earn a living as anyone else. This made servants affordable, and the middle class’s pretensions to be like the aristocracy, combined with a lack of the modern labor-saving devices, made servants indispensable of one could at all employ them.

From the detective writer’s perspective, they were enormously valuable, since they lived intimately with families to whom they rarely had any great allegiance. A brother might lie to protect a brother, or a mother her son, but there was no reason to suppose that a valet would lie to protect his master or a cook to protect her mistress. I can’t recall a single instance of anyone supposing that a charwoman would so much as j-walk for an employer.

In books, servants were not omniscient; it was possible to fool them or even to hide a body on the premises and dispose of it without their seeing. Neither were they disloyal. They would answer the questions of the police, so far as they were legally obliged to, but they did, in general, hold that repeating what they saw to strangers was no business of theirs. Discretion was important no less in maids than in doctors. However close to reality this was, it was plausible—if for no other reason than in keeping with other fiction from the timer period—and phenomenally useful to the detective writer.

The writers of mysteries has two opposing problems, and they arise out of the two principle characters of the mystery story. On the one hand, there must be sufficient evidence of the crime that the detective can detect it. On the other hand, there must be sufficiently little evidence of the crime that the murderer is willing to commit the crime at all. The near-ubiquity of servants, combined with their limitations, answer this need quite admirably, which goes a long way to explaining how frequently they showed up for the purpose.

Times have changed and servants no longer make any economic sense, outside of the homes of the unbelievably rich. The most significant factor here is that the transition in farming is mostly complete. In the United States, approximately 2% of the population are farmers; mechanization has taken its toll and the toll has been paid. Immigrants do supply a small stream of unspecialized labor, but even here the economy as a whole has developed enough jobs for people who can learn specialized skills that they do not concentrate in any particular industry. Even where they do show up in service jobs, these service jobs tend to be done on a contract basis. People no longer employ gardeners but lawn services. People rarely have maids though they may have a cleaning service. Much of the work a maid might do has been rendered doable in a short time by a washing machine, a dryer, or a vacuum cleaner. In short, live-in servants are no longer plausible. Are there any other professions which might fill the role?

I fear that, for the most part, there are not. Where people congregate they tend to pack in too closely, for the sake of efficiency, to make it easy for someone to slip something by the witnesses. Where people do not congregate, they tend to live only with people whose testimony is worthless for an alibi.

There are, of course, exceptions. Resorts will have people who work at them and at least temporarily live there, but who live in sufficiently low density that they will not observe everything which goes on. Museums, art galleries, libraries and the like also (sometimes) have approximately the right density of impartial witnesses, though they tend to be closed outside of business hours and over-packed with guests during business hours. That said, they will have slack times, of course. There are also some academic settings, such as a laboratory, that may work for the purpose, too.

All of these substitutes will have their peculiarities that will, perforce, change the stories set with them. This is no disaster, but it will make some of the spirit of the golden age mysteries harder to recapture because part of that spirit was the ordinariness that the extraordinary events took place in. One cannot make an extraordinary setting feel ordinary. Even if an volcanic observation post has the same density of impartial witnesses that a Victorian home might, it will need to be filled with the sort of odd people who might live an work in a volcanic observation post. Nearly anyone might be forced into the circumstances which make a job as a cook the only job they can get, but few people are forced by the need to avoid starvation into being a librarian. Modern writers, if we try to recapture the atmosphere of golden age mysteries, are forced to turn the characters who in the original would have been comic relief into everymen. Circumstances having changed, we must work very hard to have both the circumstances and the humanity that golden age mysteries had.