I'm still not 100% satisfied with this but I'll keep editing it forever if I don't just post it. Hopefully this provides you some of the info you were curious about (also, it will hopefully save you having to read Grafton's book. It's interesting but there're a lot of problems with it.)

A Concise History of the Footnote^1

The footnote as we know it today developed over the course of several centuries. Stemming from a long tradition of a need for citation and extra-textual discourse, the footnote quickly became the standard method upon its invention.^2

The most obvious ancestor to the footnote are the in-line citations and marginalia that can be found in ancient and medieval manuscripts. To provide an example of an in-line citation of the classical style:

Gangleri spoke: 'What was the beginning? And how did things start? And what was there before?' High replied: "As it says in Voluspa: It was at the beginning of time, when nothing was"^3,4

So while the quote comes from the Prose Edda which is a work of fiction, this example shows how the citation works. It is a call to the authority of another work to bolster the claims made by the author.^5 Marginalia are glosses written in the margins of the page, like so. Like the footnotes that would eventually develop from them, marginalia could be used for a variety of purposes including citations or commentary. In the image posted you can also see some interlinear notes. In this particular case they appear to be glosses of specific words.

Now, with handwritten manuscripts it was easy to provide external glosses. A scribe would always have room in the margins or in between the lines. When the printing press came along, it became much more difficult. The printers had to be able to typeset what they were printing. This resulted in the first commentaries being printed separate from the main text and "the first marginal annotations used in printed texts [did] not appear until 1481."^6,7 This is probably the most interesting step in the whole process, as the printers attempted to emulate the style of glossing found in the hand written manuscripts. One of the most intense ones originates in an annotated version of the Vulgate Bible known as the Glossa Ordinaria.^8 The type setting of the printed version involved "placing all the original text in the middle of the page, and creating a whole world of commentary from editor or theologians in large surrounding margins."^9 Here is a scan of a 1603 edition to flip through.

Now, the other commenter here mentions Richard Jugge as the inventor of the footnote. The main source I can find for that claim is Chuck Zerby's The Devil's Details: A History of Footnotes. The claim is actually somewhat debatable though. While he appears to be the first printer we are aware of to mark the text with superscripts tied to specific notes, his notes are actually mostly marginalia (at least in the first edition. I was unable to locate scans of later editions). The first edition that Richard produced can be found here. This is the 1568 Bishop's Bible. You can flip through and see for yourself what I am referring to.

While printers from the late 15^th to late 16^th century focused mainly on imitating the quirks of scriptural glossing, things were slowly beginning to change. In 1582 an annotated New Testament was printed; referred to as the Rheims New Testament. This work used a particular annotation style: a system of endnotes which Connors says is "one of the first usages of endnotes I have found."^10 Richard White is another early utilizer of endnotes. In the lead up to and beginning of the 17^th century, he wrote Historiarum libri cum notis antiquitatum Britannicarum. The first volume has almost 100 of its 124 pages taken up by endnotes. "The 38 endnotes…fill almost five times as much space, from page 27 to page 124, and offer a flood of primary sources to support the florid and unconvincing narrative in the text."^11 So he attempts to support his history with citations to other literature^12 but it appears that his works of choice are not up to the standards we would have now.^13

So while we have not quite yet reached the footnote, a pattern should be appearing of scholarly works appealing to sources and providing external commentary through a variety of routes. This apparent need of the scholars to source their claims continued and as we approach the 18^th century, we start seeing literary figures looking down upon scholars. Scholars are stuffy, always needing to provide evidence for everything. In a word, they are pedantic. So writers such as Alexander Pope and Thomas Swift begin to write satires of the scholarly mindset. Interestingly, the earliest English book that Connors was able to find utilizing "true footnotes is the 1710 fifth edition of Swift's Tale of a Tub."^14 It appears that the earliest example of "true" footnotes within the English canon is from an author satirizing the usage of citation and extra-textual discourse. After the appearance of these footnotes, marginalia quickly disappear from printed works:

The last purely marginal notes I have found in a serious scholarly book are in Theobald's 1723 edition of Shakespeare Restored, and Theobald was using a specific taxonomic system of Pope's editing errors that would have made footnotes less effective for his purposes.^15

So within 13 years of the "first" appearance of footnotes, they had almost completely driven out the usage of other forms of citation/commentary. Quite likely, this was fueled by the costs inherent in printing. By consolidating all of the citation and commentary to the bottom of the page, it freed up more space on the rest of the page and also made the job of typesetting somewhat easier. Consequently less space would be wasted on each page which could save a decent amount of money as the paper was the main cash hurdle for printing.^16

Before moving on, we should also talk briefly about another factor that Grafton brings up in regards to the extent of scholars citing their sources. He points to the writings and action of several philosophers and groups that brought into question various aspects of the, for want of a better term, reality that people had been living in.^17 Most notably was Descartes calling into question all knowledge besides that which is fundamental to the world, such as mathematics.^18 The usage of extensive citational glosses among scholars became a defense against such attacks on the very existence of their knowledge. The footnote was at this point arguably the most effective citation method. Yet, footnotes had not been pinned down to a specific format at this point.

Later that century, we of course get Edward Gibbon. Connors says that: "The bottom note as a literary form probably reached no higher point than it did in the hands of Gibbon."^19 This is because of how extensive yet precise his citations are, combined with his stunning display of wit in his asides. Interestingly, Gibbon did not want his work to have footnotes initially. He wanted there to be endnotes and it was potentially David Hume that convinced him (or his editor) it would be better otherwise.^20 So while he is known as a master of the footnote, it almost didn't happen.

Into the next century^21 , German scholars begin to wield the mighty footnote. They slowly began to formalize them. While discursive asides were still around, they didn't display the same wit as Gibbon. Men like Leopold von Ranke revolutionized the art of archival studies and required a solid citation format. As Ranke was a teacher of the next generation of historians, he attempted to instill in them a love of the archive and the understanding that they must point to all the archived works that they utilized. Grafton argues that providing such extensive and detailed citation had the purpose of providing a map of the process an author used to put their argument together. Yet, even with his insistence on providing extensive citation, Ranke was not a fan of footnotes. He viewed them as a necessary evil.^22

It is this gradual professionalization of German scholarship, with its competitive and even agonistic edge, that would drive the forms and methods of citation systems henceforward^23

As we finally approach the 20^th century we see the beginnings of a true codification of footnote style, culminating with the first edition of the University of Chicago Press Manual of Style.^24 Yet shortly after this codification, there started a very slow migration away from the footnote style. Endnotes were cheaper to get printed and easier to type on a typewriter. Parenthetical citations with a bibliography did not require a reader to pull themselves from the main text and made it easier to keep track of citations.^25 It is hard to say where footnotes stand. Part II of Connors "Rhetoric of Citation" shows a continuing push away from footnotes and endnotes, yet my personal experience (and I'm sure many others here) is that footnotes are still going strong. It'll be interesting to see the shift as time goes on.^26

The development of the footnote is pretty extensive—and mutilayered, as you can focus on the evolution of its design and typography, its grammatical and structural usage, or the major literary works that used it in creative new ways—but I can answer the first half of the question.

Richard Jugge, a British printer in London, is generally credited with the creation of the footnote—himself now a bit of a historical footnote. A writer with the Pittsburgh Post-Gazette suggests that Jugge moved the annotations to the bottom of the page in his edition of the Bishop’s Bible to avoid putting them in the margins, as they had been in the past. This was likely because footnotes take up only the space they need, while margins are constant, taking up space even if there’s only a few notes on the page. Footnotes made printing cheaper by wasting less space and therefore less paper.

In my research, I found a book that seems to go in depth on this very subject—The Footnote, by Anthony Grafton. If you want to learn more—and I mean a lot more, because it’s 256 pages long—you can find it on Amazon for less than $20.