One-handed operation changes everything
What a design has to give up when the second hand is unavailable.
One-handed operation changes everything because a second hand is rarely there only to add strength, it is there to hold a thing still while the first hand acts on it, and a device that assumes both hands are free has quietly delegated its own stability to a second hand that a one-handed user simply does not have available to lend.
Why the second hand does more than help
Watch two hands opening a jar and it looks like a shared effort, but the actual division of labour is more specific than that: one hand grips and turns while the other hand's entire job is to stop the jar itself from spinning along with the lid, a bracing role rather than a turning one. Remove the second hand and the turning force has nowhere honest to go, the jar simply rotates along with the lid unless something else in the environment, a rubber mat, a body braced against a counter, takes over the bracing job the missing hand used to do. Most two-handed devices lean on this same hidden division without ever stating it, one hand supplying motion and the other supplying the stillness that motion needs something to react against, which is exactly the assumption that breaks the moment only one hand is available.
This division shows up again and again once it has been noticed once: a two-handed pair of scissors relies on one hand to steady the paper while the other cuts, a two-handed screwdriver task often relies on one hand to hold the workpiece square while the other turns, and in every one of these pairings the hand doing the visible work gets the credit while the hand doing the bracing goes unremarked, right up until it is the one hand missing.
The one-handed-jar comparison
A one-handed cook wedges a jar into the crook of an elbow, against a hip, or between two cupboard doors before attempting to twist the lid, deliberately manufacturing the bracing the missing hand would otherwise have supplied. A well-designed jar opener does the same job on purpose, gripping the jar's body firmly enough on its own that a single hand only has to supply the turning force, the device itself standing in for the absent second hand's stabilising role rather than for its strength. A device built assuming two free hands never had to solve this problem, since the user's own second hand quietly solved it every time, invisibly, which is exactly why the requirement only becomes visible the moment that hand is no longer there to help.
What has to change in the design itself
A device meant to be usable one-handed needs its own base of stability built in, a wide enough footprint to resist tipping, a non-slip surface to resist sliding, or a clamp that grips the working surface directly rather than relying on a second hand to hold the whole assembly steady. It also needs every step reduced to a single continuous motion wherever possible, since a two-handed design can happily ask for one hand to hold a component in position while the other makes an adjustment, a sequence a one-handed user simply cannot perform without first setting the object down, which changes both the number of steps and the order they can sensibly happen in. Controls benefit from being reachable and operable without repositioning the grip, since a one-handed user who has to let go of the device entirely to reach a second control has lost the one hand actually holding everything together.
Sequence matters here as much as any single feature, since a task that can be reordered so that every step follows naturally from the position the hand was already in avoids the repeated pick-up-and-set-down cycle that eats most of the extra time a one-handed version otherwise costs. A two-handed design rarely has to think about this ordering at all, because whichever hand is not immediately needed simply holds the object steady in the meantime, an option that disappears entirely the moment there is only one hand to plan around.
The number that matters here
Timing the same basic task, opening a container, threading a fastener, adjusting a clamp, performed first with two hands and then with only one, commonly shows the one-handed version taking two to three times as long even once a person is genuinely practised at it, the difference coming almost entirely from the extra steps needed to manufacture stability the second hand used to provide for free. A device redesigned with its own built-in bracing closes most of that gap, not by making the single hand faster, but by removing the extra steps the missing hand's job had quietly been adding to the task.
Why this matters in practice
Designing for one-handed operation from the outset, rather than treating it as an afterthought bolted onto a two-handed design, changes which features get prioritised, a self-gripping base over a lighter housing, a single continuous trigger motion over a more elegant two-stage one, decisions that would look like unnecessary compromises to a two-handed design and look like the entire point once only one hand is actually available. The habit worth carrying forward is asking, for every step in a task, what job the absent hand used to do rather than only what it used to hold, since a design that answers the second question and misses the first still fails the moment it meets a single hand working alone.
Where this stops being true
Not every two-handed task actually hides a bracing job worth replacing, and some genuinely need two independent points of control acting at once, a task where each hand is doing something the other could not substitute for even with a perfectly designed base underneath it. Turning a steering wheel while also operating a separate lever is not really a bracing-and-acting pair the way opening a jar is, it is two simultaneous, independent inputs, and no amount of clever one-handed redesign converts that kind of task into something a single hand can do without simply removing one of the two things it was meant to control. Recognising which category a given task falls into, a hidden brace worth engineering out, or a genuinely independent second input, decides whether a one-handed redesign is actually possible or whether the honest answer is that the task needs a second hand, or some other substitute for one, and always will.