Do any other members of the primate family strike with the closed fist contacting with the knuckles? I think a chimp throws hammer fists, but that's not the same as punching...
The other apes can't make fists quite the same way we can - and coming around to a linear or hook type punch for effectiveness takes a lot more than just the wrist or foot.
Knuckle walking is not on the innermost knuckle, the way that you punch. Make a spear hand, then fold your second knucles 90 degrees in. Tense the wrist in the same way you would brace for a punch. That is a knuckle-walker's "front foot." Notice how the load is borne along the proximal phalanx, into the metacarpals that make up the palm, and from the metacarpals into the wrist structure, Which, in turn, is a remnant of our heritage as lobe finned fish. This link chain needs to be built up against the walking condition specific to the environment and habit - That is, it must be able to handle the continual load-and-unload it will be subjected to - but for the case of the other great apes, the same hand must also sustain brachiation and hand-over-hand crawling on branches, and is adapted accordingly; indeed, the wrist and finger structure of the gorillas, who spend less time off the ground shows signs of being selected for different features from that of the chimpanzee, who spends more time in the trees.
This is fine and dandy, but then the ancestors of humans did something really different. They stood up straight, and the ones who ate more and were more likely to have kids were the ones who were best at manipulating tools, requiring a more dextrous and manipulable hand, a spine, hips and back musculature that's better for standing up right than the other apes, and a foot that's meant for solely walking, rather than brachiating. So, the hand fines out and the musculature changes, but it doesn't really change more than is neccesary to the new purpose. It doesn't need to support our weight anymore, just our tools.
So now, we have two seperate 'standing' postures, and four seperate hand configurations among the great apes, the human's upright posture and tool-using hands being the most diverged from the other apes. Because of these seperate standing and moving postures, fighting at any level will be very different for a human than it is for an ape; our bodies now project their weight in an entirely new fashion. Our basic forwards step is a very specialized way of falling down and pushing up,
the other apes cannot do this. A linear, or even hook punch, doesn't make any sense in combat, because they cannot get their weight into it. Try it for yourself; get on all fours and punch straigh out. Thus, instead, they use the old James Kirk overhead special when they strike because that's how they hit harder - they can rise up easily -
which humans cannot do, as we are already there - and hammer down on something for maximum force.
My expectation, then, is that the human punch is a secondary adaptation of several other human traits - it combines the hand, and wrist structure of our knuckle walking ancestors with our current upright posture with the shoulder, hip, knee, and ankle structure of our upright, tool-using ancestors, and in doing so may provide a limitation against a more gracile structure, rather than being a structure novelly adapted for the strike.
I'm with you there: man's hands were made for clubs.
But you kinda lost me there: the spear's only about 350,000 years old.
A lot of the general tool use adaptations of the shoulder and wrist go into throwing - it's a behaviour that once again, adds up many other little things, but it's very different from a stored energy missile weapon.
