There are now tens/hundreds of millions of young children all over the word why are being given their very first tablets and even phones to play with at a very young age.
Perhaps even 2 or 3 years old for a tablet to play with.
As far as this whole generation of young children are concerned screens are tactile items to be touched.
Yes, when you learn to write things you can use the keyboard for letters/words, but you are use the screen to swipe pictures, scroll around, zoom in/out.
It's a whole package, not one or the other.
These children will easily have the first 10 ish years of their lives, from the moment they remember being able to physically interact with a screen.
So, at the age of what..... 11 or 12, you going to say to them, ok, enough of that, here's a real computer, and no you can't do anything with the screen on a real computer.
You're assuming that all levels (or classes) of task can fall easily under one method of input and that should be the 'dominant' one. In reality different methods of input are best suited to different tasks - and if users seek the best approach, they will use the one that's appropriate to that task.
For instance, you can't just tell a mechanical engineer to switch to a touch interface when using Fusion360 or AutoCAD because "touch is better for everything else, so get used to it". Likewise, you wouldn't tell an animator that they can only use a mouse and keyboard when their work is designed around a pen tablet.
The primary concern isn't about which method of input is
better, but how they fall into
different classes of device.
Note the word 'device'. To this point, Apple has actually been very smart in holding back from a convertible/all-encompassing touch/trackpad/keyboard device. Like other markets, they've observed the landscape and analysed what consumers want versus what they need.
Other manufacturers have experimented for many years now and used consumers as guinea pigs for their different ideas, and we've had a lot. Rotating screen, folding screen, screens that rotate back on themselves, detachable screens... But there has yet to be a single solution that satisfies the majority of users, and inevitably support for this hardware gets abandoned quickly. And that is the biggest problem, that armchair critics continually think that there has to be 'one' solution for everyone.
Microsoft may well be the business lecture example of this in years to come. They sought to create the 'single' solution with Windows 8, but the irony was that they actually fragmented their user base more because OEMs manufactured a range of devices with no common ground - and the software, while trying to the Jack-of-all-trades, was the master of none. With Windows 10 they reverted back to a more 'classic' Windows UI, which thus alienated the touch-crowd.
Likewise their Surface tablet has an entirely different concept to the Surface Studio, despite being branded under the same banner. And fewer Windows-based devices are now being sold with touchscreens because the market segment that
does want touchscreens in a laptop-like device will use ChromeOS, Android or alike.
Where does this leave Apple? Their solution right now is the iPad, which can be paired with a physical keyboard and trackpad. The complaints against this setup are that it becomes expensive, heavier than a MacBook Air and that iPad OS doesn't have the same level of features as macOS - all valid arguments.
But
if - and I emphasis if - Apple feels there is an opportunity to market a device that can be used with a trackpad, keyboard and touchscreen, and which can combine the simplicity of iPadOS with the more fleshed-out experience of macOS, it will almost certainly come in the form of a new product category that is neither an iPad nor a Mac. Both of these product categories were designed around specific input methods and Apple is smart enough to know that they can't force change on either.
In short: macOS and Mac hardware is not designed for touch input.