It seems like many manuals and technical documents prefer passive voice over active voice. Is this true, or is it just my perception? If so, why?
I can add that more than a decade ago, I was senior tech writer with an engineering & manufacturing firm whose product was used in semiconductor fabs, and when I started, the vast preponderance of their tech docs were littered with egregious passive voice, as well as massive redundancies and an over-reliance on repeated heavily-bolded all caps warnings where there were use-case risks.
I found, after discussion with the heads of Document Control and Engineering that this was a result of the one class each engineer had been required to take in tech comms during university - and that independently, at no less than ten top-flight US engineering programs, these student were taught that passive voice would somehow help them magically evade liability and responsibility should an enduser be harmed in the use of their product.
Needless to say, this is a baseless belief, belied by both competent writing and many years of case precedent, and once I presented studies, and brought in a risk-specialist attorney to explain this, we shifted language.
"Should an incompatible gas be flowed through this purifier's getter bed, an uncontrollable runaway exothermic reaction may occur, with risk of harm to nearby persons or property. When facilitizing this unit, confirm gas compatibility."
"Caution: this purifier may explode or catch fire if the wrong gas is flowed through the getter bed. Double-check for correct gas to purifier chemistry at installation and again prior to flowing gas."
I suspect that though this was more than a decade and a half ago, the same thinking may still apply.
In academic writing, scientific papers and similar, it is normal practice to avoid writing in the first person. Such documents are written in the third person. Check the writing style guide of any scientific journal for confirmation of this.
This practice carries through to a large number of technical communications. When writing in the third person, the passive voice becomes more natural. There is a tendency to follow on from:
The system does x
The system is affected by x
x affects the system
This tendency doesn't make the use of passive correct or preferred. It is just a common tendency among some writers.
It would be recommended to continue to use the general principle of minimising the use of the passive voice.
Speaking only from my own experience, passive voice is often a last resort. I write a lot of user guide material, and my company's style guide discourages using passive phrasing, but even more strongly discourages using second-person pronouns. So while it may feel more natural for me to write "You can access the setting in the widget," I may have to use "The setting can be accessed in the widget" instead, assuming I can't figure out another way around it.
I suspect this is true of other widely-used style guides as well.
Well, there is no good practical reason for it. In other words, there are no studies showing that passive voice is more effective in communicating technical information. That leaves us with social reasons, which are necessarily a little more speculative and anecdotal. There is history to suggest that the following may be factors:
People often feel that because documentation is a "business document" it must be formal, and they often feel that the passive voice sounds more formal than the active voice. Neither of these propositions is true but they seem to be frequently held by people whose training and experience is not in communications.
The passive voice can be used to conceal the actor. "John hit Frank" becomes "Frank got hit" with the accusatory "by John" elided out. Thus it absolves the speaker either of taking responsibility or of assigning it. Communicators who seem unsure of themselves or who don't really want to communicate at all often take refuge in the passive to avoid taking or assigning responsibility.
If people are in a field where passive documentation is common, they may simply copy that style. It is remarkable the extent to which people will produce information designs that they would never use or like themselves just because they think those designs are conventional. People choose the conventional over the useful all the time.
I don't think it's a default standard any longer, and hasn't been for some time. Perhaps it's just that older documentation is still hanging around or used as a base document for similar, new products without any thought (or time) to update it to active voice. Personally, I've avoided using passive voice as a default for the past 20 years (and surreptitiously updated an entire suite of Standard Operating Procedures to active voice over the course of a few years).