Well, the first thing you'll probably notice is that the sun rises every day and sets every night. During the daytime, when the sun is up, the sky can get pretty bright. You can't see much up there except clouds and blue. At night, when the sun is down, the darkness allows you to see more lights in the sky: the moon (also sometimes visible by day), the stars, and the planets. But let's set those aside for now.
What can the sun, coming up and going down day in and day out, tell you about the passage of time? The first thing you'll notice is that there's one day after another, each one followed by a night. Day and night, day and night, day and night. Not much good. It soon gets harder and harder to keep count of them. You're going to have to invent some pretty big numbers if you're going to go on that way.
But soon you become aware that not all days are the same. The days, on average, seem to get hotter and longer for a while, and the nights between them proportionally shorter. Then for a while the days will grow shorter and cooler, and the nights longer. You might be able to confirm this with some precision by inventing hourglasses or striped candles or something like that. But really, that's an awful lot of technology to be creating at such an early hour of the day, historically speaking, when you're not even sure how to tell what day it is. Glass blowing? Pffff! Fat rendering and dyeing? Not likely.
What you're more likely to do, I imagine, is figure out a way to measure the angle of the noonday sun above the horizon, or the length and direction of shadows at given times of day. All you need for the one is a notched stick or a knotted string, and some kind of sundial for the other. The latter could be invented using a tree stump and stones to mark the position of the shadow.
If you're going to get anywhere with this experiment, though, you will need to perfect some form of record keeping. It doesn't have to last forever, though no one will know you did it a few thousand years hence unless it does. What will you invent? Pen, ink, and paper? Paint on a plaster wall? Embroidery on woven fabric? Engravings on wood, stone, or metal? Maybe mosaic tile? Choices, choices! And will your records be simple pictorial representations, or will you have to squeeze out a system of written symbols as well? What tools will you need to ensure that your measurements and records are accurate? What a lot of trouble! But you can't make a stew without slaying a mammoth, so...
All right, now you've documented the fact that the days grow longer, and then shorter again, turn and turn about, in a regular cycle. It's always so many days between one equinox and another (when the days and nights are of equal length), and the same number of days between solstices (when the difference is as big as it gets). Maybe you're not so sure about your measurement of the lengths of the days and nights, but you've got all that shadow and noon-elevation data backing you up. The sun rises and sets farther and farther to the south, then starts coming north again. The noontime crest of the sun's arc gets progressively lower, then starts getting higher again, within predictable limits and between predictable numbers of days. The prevalence of hot and cold weather seems to vary along the same schedule. So you've got summer, autumn, winter, and spring; solstices and equinoxes; 365 days divided into observable and measurable units. Voila! You've discovered the year!
Maybe it happens that way. It would be especially likely if the scientific geniuses studying the phenomena were morning people. But what if they're night owls? Then maybe it happens this way instead:
Memory being what it is in preliterate cultures, you will soon observe that most of those night-lights continue to shine in the same apparent shapes. Though a given constellation, and indeed the entire night sky, will seem to turn slowly around your fixed viewpoint from night to night, until it goes all the way around, the stars remain fixed in the same patterns, unmoving in relation to each other. Otherwise you wouldn't be likely to notice the way the sky turns full circle in a given number of nights that also, coincidentally, corresponds to the seasons observed by your neighbor the morning-person.
From said morning-person you might steal that notched stick, or knotted string, or the idea for it, or something similar. Then you can make even more accurate measurements of the angles between the stars in a given constellation, and the apparent distance between the constellations, and the direction from which they rise and set each night, and how that direction changes from night to night. You will notice some rising higher or lower than others, some rising to different angles above the horizon at different times of year, and some disappearing altogether at certain times of year. But by now you must have a notion of a year because the cycle keeps repeating.
But you're also bound to notice that some of those night-lights don't know their place or stick to it. Besides the sun and moon, there are star-like lights that meander around, breaking the pattern of fixed lights in a night sky that turns around the earth. Investigating the movements of these planets may mean inventing more accurate measuring tools. But it may be as simple as assigning territories to each of the constellations, and plotting out on some kind of map the path that the planets take as they move from one constellation to another. Ditto with the rising and setting of the sun and moon, which will take place against different constellations in the stellar background as the year goes round.
It may take precise instruments and finely-tuned drawing and writing techniques to figure out that the sun, moon, and planets each follow a pretty consistent path through the Zodiac (i.e., the constellations whose territory lies along the curve of their apparent path through the night sky). And then, gosh darn it, you've discovered the ecliptic plane. And if algebra doesn't necessarily come along for the ride, at least geometry must be discovered and advanced to a certain point.
These discoveries and inventions seem to happen repeatedly, and sometimes simultaneously, in different cultures around the world. Sometimes they are forgotten and rediscovered. Sometimes the lore of a previous discoverer is passed down through generations, and even across cultural boundaries. Of course the science is often mixed up with all kinds of magical and religious twaddle. But the occult and cultic caliber of this knowledge is the reason it will tend to be preserved in a medium that will last long enough for future centuries to unearth it and marvel at the advanced state of ancient astronomy.
This suggests that a lot of discovering and inventing goes into recognizing that there is such a thing as a year. Yet all of it must happen right away in human history, because history is nothing without the ability to count years. Did somebody tell the first would-be astronomer to be looking out for certain patterns, and to prepare to be amazed at what they see?
Also, let it not be forgotten that certain developments in culture must take place before anyone is likely to have the leisure to make mystical calculations about the movements of heavenly bodies. I mean, anybody around the campfire can tell stories about gods, monsters, and heroes who personify the lights and groups of lights in the night sky. But somebody first has to notice that there are such groups of lights and observe how they are and aren't moving from night to night. And that person is going to need a good excuse to be exempted from hunting and food-gathering duties if he's going to have the time and wakefulness to make these observations. This suggests a strong likelihood that before a culture progresses as far as reading horoscopes, it will need to achieve refinements in agriculture, technology, and economics that make room for an idle class of priests, wizards, astrologers, and alchemists, among other useful callings.
So while the invention of the year might go to the credit of a wide-awake hunter-gatherer, chances favor the honor going rather to nomadic herdsmen, or citizens of an agricultural village, or somebody in the streets of a fortified town where such a chimera as a merchant may be heard of. They might be smearing their discoveries on cave walls with paint made from inedible plant and animal fluids, or digging them into stones with fragments of sharper stone. But if we're going to find out about it, they will probably have to have invented mineral dyes, or mosaic tiles, or refined metals with which to chisel sculptures in wood and stone.
The fact that cold and hot weather come in cycles may be of use to people whose livelihood depends on their ability to migrate whithersoever the wild-growing nuts, fruits, grains, and vegetables happen to be growing, or prey animals browsing. But counting years will probably only begin to matter around the time when some grain merchant realizes that he needs to invent compound interest. Unless—and here's the obligatory plug for intelligent design—unless the invention of the year was a gimme, a freebie from the Maker as he handed over the world to its first human stewards. "Before you take over management," He might have said, "there are some things you ought to know..."
