Why do you assume that it didn't? There's plenty of emerging evidence that even foraging humans had a fairly complex social organization, with "proto-cities" that functioned as ceremonial centers at the very least. That's circumstantial evidence that they did have "civilization" of a sort, and perhaps even "modernity".
Just because the neolithic period started 12-15k years ago doesn't mean somebody woke up one day with a modern brain smart enough to plant things.
1. Hunter gatherers collect seeds, but drop some near their temporary settlement.
2. The next time they stay in the area, they notice food plants growing where they dropped the seeds, so they don’t have to travel as far to collect them.
3. They start deliberately spreading seeds so there will be even more there next time.
4. They start clearing land to make more space to spread the seeds on.
5. The quantity of seeds they can gather at the site meets or exceeds their needs, so now they don’t need to travel to gather enough and can settle down.
Throughout the process they discover little tricks, like scratching the ground before scattering their seeds, or which kind of ground is best to clear and sow, which seeds grow best, etc, etc. This could take many generations but once the process starts under the right conditions with the right seeds in the right area it snowballs.
Some languages such as Guniyandi have only recently undergone this conversion , in fact there are likely thousands of languages still without writetn forms 
Figuring out that it could be used to write sentences took a few millennia after that. And only after that came enough simplification for people other than palace/temple employees to learn it.
"The original Sumerian writing system derives from a system of clay tokens used to represent commodities. By the end of the 4th millennium BC, this had evolved into a method of keeping accounts, using a round-shaped stylus impressed into soft clay at different angles for recording numbers. This was gradually augmented with pictographic writing by using a sharp stylus to indicate what was being counted. Round-stylus and sharp-stylus writing were gradually replaced around 2700–2500 BC by writing using a wedge-shaped stylus (hence the term cuneiform), at first only for logograms, but developed to include phonetic elements by the 29th century BC. About 2600 BC, cuneiform began to represent syllables of the Sumerian language."
The monumental inventions are:
3. printing press
because they are ways to store and disseminate knowledge.
This is sometimes called culture.
Not just that. But someone had to invent:
* Pottery (Lets store food, instead of looking for it all the time)
* Cities (Hey, lets all settle here and come back here for some reason)
* Basic tools: Rope, Hammer, Wedge, Knife. Metallurgy.
If we were trapped in a hunter-gatherer society, even with our advanced brains today... it would take a very smart person to invent a city and/or writing. And you'd need the support of the entire hunter-gatherer society to bring you food and resources to fund the inventions.
Cities invent the aristocracy, a group of people who are dedicated to advancement of culture and technology. Agriculture and Pottery provide a way for transportation and storage of resources (food especially), supporting the aristocracy.
At which point, true inventions can begin. You can't invent new tools if you're too busy hunting for food every day of your life.
Even the spear-thrower, followed by the bow and arrow, took a loooooong time to invent.
Even if the bow was invented, creating good fetching and requiring the mass production of fragile arrows makes it a poor weapon in the Dark ages.
You pretty much have to invent mass production + commodities, standardizing the arrowhead before Bows become superior to spear and/or axe throwing. Once "any" arrow can work on "any" bow, you are golden. But this commodity idea is very difficult to come up with.
Shoot an arrow into a deer, and the arrow is ruined. You need to build a new arrow. Throw a javelin or Axe into a deer, and you can use the same Javelin / Axe on the next deer as well.
1. arrows are reusable. I've dabbled in archery.
2. arrows are a distance weapon. You can take out prey or enemy at a distance. Throwing an axe is very short range.
3. throwing an axe means you get one shot. Then you're weaponless.
4. throwing a metal axe is much more effective than a stone axe. I seriously doubt throwing a stone axe is very effective - you'd be better off just throwing a rock.
Romans carried 2 Pilum, a shortsword, and a shield. You had two shots, after which you rushed in with your swords.
Yeah, Longbowmen are obviously superior. But if you have 10,000 Longbowmen with 60 arrows per man, you require a production-capacity of 60,000 arrows per battle. That's a lot of goose feathers, even if you're recycling a significant chunk of arrows.
Note that the modern arrow has fletching for accuracy. Ancient arrows didn't have fletching: no spin, no accuracy.
Before the middle ages, mass production of arrows in this capacity was basically impossible. The British innovation to war wasn't so much the Longbow (there were plenty of archers before the British...), it was the invention of commodity mass produced arrows... which supported the Longbowman as a major unit of warfare.
Even the Roman Legionaries used Pilum (aka: Javelins) as their ranged weapon of choice. Bows existed (see the Sagittarii), but were an auxiliary (and often non-Roman) force.
Slingers (!!) were still used in Roman times. Now a Sling... THAT is an ideal ancient weapon. A trained slinger can kill a Lion or Bear, and you only had to find rocks to throw. Mass production of sling-bullets (really, just a shaped rock) is easy, even with an army marching to a distant location.
Slingers used rocks, clay, and even lead bullets throughout history. They were effectively the longbowmen before the British learned to mass produce arrows.
In fact, the ancient world believed that Slings had greater range and accuracy than bows. Since fletching won't be invented for another 1000+ years, ancient arrows had very limited range and accuracy.
I couldn't find any mention of when spin was added. But I would think it pretty darned hard to align the feathers so precisely that it wouldn't spin anyway. Bird feathers aren't straight, either.
Arrows date back 64,000 years (wikipedia). A long time before the Romans.
Equipping 10,000 soldiers with any sort of kit will be a major undertaking.
Most any bird feathers can be used for arrows. Not just goose.
Difficulty of that kit is the real question.
Slingers would need 20 lead bullets, one of the lowest melting point substances in the world. If lead were unavailable, slingers would make due with stone, or even clay bullets.
In contrast, equipping 20 arrows to 10,000 archers would be an incredibly more complicated undertaking. You need arrowheads, those arrowheads need to fit shafts. Even without fletching, building an arrowhead design that fits different width sticks of wood is an incredibly complicated undertaking. Especially if you only have access to tools from ancient-times.
To accurately launch the arrow, you must consistently make the same arrow (or extremely similar arrows). That's just how it works. Its far easier to make 20x lead bullets (or 20x clay bullets, or 20x stone bullets) that are all consistent... than it is to make 20x arrows that are all consistent.
It was a real innovation to standardize stick sizes, standardize goose feathers, and standardize arrowheads to equip the British Longbowman. There's a reason why that strategy only really took place in the Medieval period, the ancient world didn't have the tools or inventions needed to support mass-archer strategies.
Sorry, not agreeing. Arrows had been regularly made for tens of thousands of years.
What if you had two axes!
Progress is incremental. One step after the other.
If I was dropped into Roman times, what I'd invent is a printing press. After inventing paper made from rags.
This question is part of what inspired Harry Binswanger’s book, “How We Know”:
“Mankind has existed for 400,000 years. But 395,000 of those years were consumed by the Stone Age. The factor that freed men from endless toil and early death, the root cause of the elevated level of existence we now take for granted, is one precious value: knowledge. The painfully acquired knowledge of how to master nature, how to organize social existence, and how to understand himself is what enabled man to rise from the cave to the skyscraper, from warring clans to a global economy, from an average lifespan of less than 30 years to one approaching 80.”
Finally, infant mortality was extremely high and remained very high until modern medicine. These infant deaths decreased the average life expectancy but it masks the lifespan that healthy adults would typically experience.
You can argue that those who survived infections, the elements, violence and starvation were in much better shape to survive everything else.
Also contrary to popular belief, cardiovascular disease happened before modern times, and even if we might be talking of Neolithic only, this happened even in hunter gatherer populations. Found these pretty interesting:
In modern times what we have is much less violence, vaccines, antibiotics, a healthcare system capable of healing injuries or restore you after heart attacks, etc and in spite of prevailing beliefs we’ve got proper nutrition and at least in the industrialized countries we aren’t dying of starvation or nutritional deficiencies anymore.
Our problem is “energy poisoning”. Once we figure out how to not eat so damn much, we’ll eliminate most chronic disease we have today.
An aspect I found particularly interesting was the idea that basic physical geography could have played a strong role in determining when and where complex civilizations arose.
(title: African History Disproves “Guns Germs and Steel” by Jared Diamond).
tl;dw: that theory is not as valid as it may seem at first.
In order for the mind to conceive of something, first there must be the idea that it is possible. For example, in the 1970's lots of people wondered what a home computer would possibly be good for. The only things people could imagine were storing recipes and balancing one's checkbook. I'm not kidding.
There is a theory that humans only developed consciousness in the last 2-3000 years, based on literary sources, and that we were sort of guided by what were in fact auditory hallucinations (generated by another, less connected part of our brains) before that time.
> “It is one of those books that is either complete rubbish or a work of consummate genius; Nothing in between! Probably the former, but I'm hedging my bets.”
Even if it’s totally wrong, it is an interesting thought experiment.
I keep hearing this, and it's probably true in some sense.
But agriculture is also an incredibly good deal, in that it now supports 7000 million human lives. H&G only reached about 50 million.
So it becomes a philosophical question: Is it better to have a few happy people alive or a huge population of less happy ones?
I'm reminded of the old "explorer's fallacy", where you encounter a tribe in the jungle and marvel at how healthy everyone is. Not realizing that the sick and infirm aren't allowed to live there.
I think there are many studies on the overall health of hunter-gatherers as a result of diet and exercise and leisure time as demanded by the lifestyle. In early agricultural societies you have to work harder and longer and you're forced to eat a lot of bread, potatoes and corn and other low quality foods.
> H-G numbers were kept in check by available resources randomly dispersed in nature.
But... doesn't "numbers were kept in check" mean "surplus people starved to death"?
Not so obviously. Would you argue this for (say) lions or eagles? Few adults starve, when times are lean there is less breeding, and fewer kids who survive, compared to times of plenty.
Whereas the collapse of an early agricultural society (perhaps due to pests, or just no rain) probably did result in lots of adults starving, if there were more of them than could be supported on gathered food.
In the sense that the diets & health of ordinary people were much worse under early agriculture. This appears to have been true for a long time, maybe until a few centuries ago.
People ran away if they could. Early agricultural societies seem more and more like attempts by warlords to enslave both animals and humans for their benefit. Capturing livestock & farmers seems to have been the main goal of warfare.
> also an incredibly good deal, in that it now supports 7000 million human lives
Sure, 21st century life is (for almost all of us) much better. It's not so clear that 19th C life was so (except for a few highly-developed places).
TL;DR: there are trade-offs.
edit: I forgot to mention, there may be a third way that combines the best of both worlds: "Redesigning Civilization with Permaculture" (video) http://tobyhemenway.com/videos/redesigning-civilization-with... sequel to "How Permaculture Can Save Humanity and the Planet – But Not Civilization" http://tobyhemenway.com/videos/how-permaculture-can-save-hum...
Basically as the population grew, hunting and gathering became insufficient to feed the population.
Also bone analysis shows that the health effects you mentioned happened probably BEFORE widespread adoption of agriculture, possibly due to malnutrition.
This isn't far-fetched in an environment in which people are starving.
And with the adoption of agriculture it got a little worse before it got better. Height and life expectancy started going up again.
Basically we built better tools, we started diversifying the crops, we learned to cope with the nutritional deficiencies (for example we started eating dairy) and we eventually adapted, like we always do.
And of course without agriculture it's very hard to form a lasting society. People move around too much. Without society it's very hard to keep clever inventions from dying with their creator.
Relying on hunting and gathering was also risky: A harsh winter would decimate your prey, and then you would starve.
I guess it's hard to have numbers, but early attempts at agricultural states (often in these same locations) seem much more fragile. Most collapsed within a few generations, to disease of humans or crops, or to weather, we can't know.
Neither did early agriculturists.
> early attempts at agricultural states (often in these same locations) seem much more fragile
That's because there are no hunter-gatherer states. Hunter-gatherer cultures are much too fragile to support the societal structures needed for even primitive states.
And to make sure it worked to your benefit (as the king) it was important that they farmed a limited number of things which were easy to keep track of, and easy to transport, and hence easy to tax. Like wheat or barley or rice, which you can inspect any time, and which ripens on a predictable date, on which your soldiers can show up to collect the rent. (And unlike potatoes.)
The comments there are generally of high quality, and you'll find some good counterpoints to the book's claims.
Counterpoint: "someplace better" might already be occupied by other humans. Besides, a harsh winter is not a local phenomenon, it affects entire regions. Good luck getting to "someplace better" if you're starving, don't have any transportation besides your feet, and don't even know where "someplace better" even is.
Early civilization was superior to a hunter-gatherer lifestyle in only one aspect: how large of a population it could support. In nearly every other way, it provided a worse lifestyle.
From TFA it appears that the original human population in Africa was pushed out of the interior by climate changes, it's not inconceivable that those conditions also provided evolutionary pressure
Some people have posited that it's because some fundamental biological limits were run into. (Our biology has already made many compromises to support the large brain, such as extended childhood, compromises in women's skeletal structure, the large energy consumption of the brain, etc.)
This would have significant implications for potentially different cognition among different human subgroups. Especially those that have lived in isolation for 10s-100s of thousands of years until recent history with modernized travel and interconnectivity.
This may not be a pleasant realization, but it cannot be overlooked in the quest to understand sociological dynamics.
 https://www.biorxiv.org/content/10.1101/166520v8.full, 2019
The comment in question: https://www.reddit.com/r/AskAnthropology/comments/4ehwdp/wou...
I do recall, on hacker news I believe, an article posted based on a study by NASA scientists that tried to calculate the likelihood of finding evidence for civilizations 10 thousand years, 100 thousand years or older. What is shocking is how quickly any remnants of an advanced civilization could deteriorate. IIRC, the conclusion of the study was that the chance of us finding any evidence of civilizations that old were extremely small.
The idea that there was an ancient civilization over 10 thousand years ago is entirely plausible to me. Massive geological events (or even astrological ones like asteroid impacts) could have altered entire continents. Who knows what evidence is at the bottom of the Pacific ocean that we cannot access. Or what evidence has been forced under tectonic plates through subduction.
Of course, extraordinary claims require extraordinary evidence and so far none of the presented evidence is extraordinary (IMO). But our fixed notion that civilization sprang up nearly fully formed 10 thousand years ago doesn't seem as clear to me as it once did.
I can't be sure this  is the original article I was remembering but it might be an interesting read anyways. I am sure the study authors approach the subject with far greater care than I could.
Concrete, on the other hand, is advanced.
The novel "The Hab Theory" is a good story about the discovery of prehistoric civilizations.