Three levels of claim
Every statement on this site belongs to one of three levels, and the wording tells you which.
Documented. A surviving source says it, and the source is named: a letter, a draft, a printed text. Example: Leibniz's 1679 list of people to write to about a society begins "Helmont; Rosenroth; Morus" (A IV,3 N. 133). That shows whom he meant to approach. It does not show that any of them joined anything.
Inferred. The documents fit together in a way that suggests a link, but no single source states it. These sentences say so in plain words, for example "the editors conjecture the date" or "this is my reading".
No documented membership. Where no record shows that Leibniz or one of his correspondents belonged to an order or society, the site says "no documented membership". It never says "was not a member". Societies of the seventeenth and eighteenth centuries were often discreet, some members denied belonging, and silence in the record is not proof of absence. The question stays open.
About the catalogue
The catalogue lists every piece in each collected volume of the Akademie-Ausgabe, and shows them in two ways: in order of date, and volume by volume.
- Titles and dates are the editors' own, mostly in German, read by machine from each volume's contents pages. In the letter series "an" is given as "to" and ranks and month names are put into English. A few titles may be misread or cut short, and some volumes are marked as incomplete.
- Each piece is placed in the year the editors give it. Where they give a span, such as 1686 to 1689, the piece sits at the first year and the span is shown beside it. A question mark is the editors' own doubt.
- An enclosure with no date of its own takes the date of the letter it was sent with. Any other undated item in a letter volume takes the place of the piece before it, since those volumes run in order of date, and is marked "placed by position". Writings the editors leave undated are kept apart as undated.
- The six periods on the timeline are the usual divisions of Leibniz's life. They are a reading aid, and nothing depends on where one ends and the next begins.
- For the original of a catalogued piece, follow the link on its volume page to the edition itself. Originals are printed on this site only beside a translation made here.
- "English here" marks a translation on this site. "English in print" names a published translation. A piece with no mark has no English version located yet, and most pieces have not been checked. It does not mean that none exists.
- Volumes that exist only in print, Series V, the older editions and unprinted manuscripts are not yet itemised.
Symbols used on this site
The visual marks on these pages come from Leibniz's own work. They are design devices that point back to him, and they make no historical claim on their own. Each is documented in his writings or letters. The principal works they come from are being translated for this site and will be linked from here as they are added.
| Mark | Where it appears | Source in Leibniz |
|---|---|---|
| The point | The moving fields of points behind each page. On the home and catalogue pages each cluster is one period of his life. | The monad, a simple substance without parts in which the whole universe is mirrored from one point of view (Monadology, 1714, sections 1 and 56). |
| Binary numbers | Each text's number is also written in zeros and ones. | The dyadic arithmetic, "Explication de l'arithmétique binaire" (written 1703, printed in the Paris Academy's Mémoires), translated here. |
| Hexagrams | The six-line figure beside each text is its number in binary: a whole line is 1, a broken line is 0, lowest digit at the bottom. | In 1701 the Jesuit Joachim Bouvet showed Leibniz that the hexagrams of the Yijing in the Fuxi order can be read as binary numbers, and Leibniz adopted the reading. The figures here are numerals only; no divinatory meaning is intended. |
| Sums and differences | The field on the home and catalogue pages. Each cluster is one period's share of the surviving pieces, the curve through them is the running total, and the area under it is built from thin upright strips. | Leibniz's calculus treats a curve through its differences and the sums of them, each undoing the other. He published the differences in the "Nova methodus pro maximis et minimis" (1684), translated here, and the sums, with the long S that became the integral sign, in "De geometria recondita" (1686). |
| The long S, the triangle and the strips | The sign before each total on the home page, the small triangle in the corner of each panel, and the rows of thin upright strips under panels and above sections. | The long S for a sum and the d for a difference first appear in his Paris notes of 29 October and 11 November 1675 (A VII,5). The characteristic triangle, whose sides are the differences dx and dy, is the figure he said opened the calculus to him. The strips are those of a quadrature, as in the "Supplementum geometriae dimensoriae" (1693). His calculus writings are listed on the calculus page. |
| Sphere and plane | The field beside each translation: a sphere of points around the year, two marked points, and a ring of points lying midway between them. | In the "Characteristica geometrica" of 1679, and in the essay on it that he sent to Huygens that September, Leibniz defines figures by the situation of points alone: a sphere is the place of all points that stand alike to one point, a plane the place of all points that stand alike to two. These pieces are not yet translated here. |
| The wheel of combinations | On the translations page the themes sit on a ring, and a chord joins two themes when one text carries both. | The diagram of combinations at the front of the Dissertatio de arte combinatoria (1666). |
| The medal | The round emblem on the home page, and the ring with a stroke beside the site's name. | At New Year 1697 Leibniz proposed to Duke Rudolph August of Wolfenbüttel a medal showing the numbers written with 1 and 0 as an image of creation, with the legend Omnibus ex nihilo ducendis sufficit unum, "to draw all things out of nothing, one suffices". The drawing here is new and copies no engraving. |
| The dyadic table | The field of points on the Ask page, and the small numbers in zeros and ones beside headings. | The table of numbers in the "Explication de l'arithmétique binaire", translated here. |
| The mirror of the whole | The strip under each translation: every year of Leibniz's working life as a bar, with that text's year lit. | Each monad "is a perpetual living mirror of the universe" from its own point of view (Monadology, section 56). Each page shows the whole body of writings from where it stands. |
| No leaps | Pages cross-fade into one another, in browsers that allow it. | The law of continuity: no change happens by a leap. Leibniz states it in Part II of the Specimen dynamicum, 1695, paragraphs 4 to 6. |
Three mottos take turns in the footer. "Theoria cum praxi" is the one Leibniz gave the Berlin Society of Sciences in 1700. "Calculemus", let us calculate, is his word for how disputes would be settled once reasoning could be written in a universal notation; it names the Ask page. The third is the legend of the medal.
How the points move
The points behind each page are moved by rules taken from Leibniz's dynamics. This is a design device, as the symbols above are, and it makes no claim to be a physical model.
| Rule on this site | What you see | Where it comes from in Leibniz |
|---|---|---|
| Living force is conserved | Every point has its own velocity, and the sum of the squares of all the velocities is held at one fixed amount. When your pointer drives some points hard, the rest slow a little, and the large motion is soon shared out among the many. Nothing is added to the field and nothing is lost from it. | Leibniz measured force by mass times the square of the speed, his vis viva or living force, against the Cartesian measure of mass times speed, and held that the same quantity of it is always kept in the world: the "Brevis demonstratio" of 1686, the Discours de métaphysique, section 17, and the Specimen dynamicum, Part I (1695). |
| Dead force becomes living force | The pointer never moves a point directly. It only presses on it, moment by moment, and the pressure builds into motion. Each point is also held to its place as if by a spring, so it swings back and past. | The Specimen dynamicum, Part I, distinguishes dead force, a mere solicitation such as weight or the tension of a spring, from the living force that arises from "infinitely many continued impressions of dead force". Part II argues that every body, however small, is elastic. |
| Nothing happens by a leap | No point jumps and no light switches on at once. Positions, speeds and brightness all pass through every degree between. A fast point leaves a faint trace of the small step it has just taken. | The law of continuity, which Leibniz stated in a letter printed in the Nouvelles de la République des Lettres in 1687 and used in the Specimen dynamicum, Part II: "no change happens by a leap". The small step is the difference of his calculus, as in the Nova methodus (1684). |
| Harmony without influence | No point acts on any other. The band of light that crosses the field is not passed from point to point. Each point brightens by its own clock, and the clocks were set once, at the start, to agree. | The pre-established harmony of the Système nouveau (1695), and the image of two clocks that keep time together because they were made to, in the postscript to Basnage de Beauval (1696). |
| Each point mirrors the whole | Every point on the page answers your pointer wherever it is: plainly when it is near, and by an amount too small to see when it is far, but never by nothing. | "Every body feels everything that happens in the universe", and each monad represents the whole, distinctly in what is near and confusedly in the rest (Monadology, sections 56, 60 and 61). |
If your device asks for reduced motion, the points stand still.
Asking, listening and dictating
The Ask page reads a question for names, years, kinds of piece and subjects, searches the catalogue titles and the English of the translations, and states what it finds in a sentence with a chart. That part is plain counting. Subjects are matched against a short list of the Latin, French and German words that titles use, so a subject no title names will not be found.
Where the site's server offers it, an open-weights language model, Qwen, then writes the answer. It runs on the server, so nothing is downloaded to the reader's device. It is given only the question, the counts and the passages found, and is told to cite the passages by number and add nothing from memory. A model can still misread a passage, so the passages are always shown beneath the answer. When the model cannot be reached, the counted sentence is shown instead and says so.
Every translation and every page of prose can be read aloud, and search boxes take dictation where the browser allows a microphone. On the site's own address, dictation is done by Voxtral, an open speech model from Mistral that runs on the site's server: a recording of at most about twenty seconds is sent when you stop speaking, turned into text, and not kept. On the site's own address the English translations and the Ask answers are read by a Speechify voice, made on the site's server from the site's own text; anything else, and everything when that voice is not connected, is read with the voices of the reader's browser or device. No browser has a Latin voice: Latin originals are read with an Italian voice, which is the nearest available and is not a reconstruction of how Leibniz spoke.
About the translations
The translations are working translations made for this project with AI assistance. They have not been peer reviewed. Each one is printed beside its original so that it can be checked line by line, and corrections are welcome.
- Each text gives its edition reference, for example "A IV,3 N. 130" for Akademie-Ausgabe, Series IV, volume 3, number 130.
- Line numbers and line-end hyphens from the edition are removed. Spelling is otherwise as the editors transcribed it, including slips in the online text.
- The 1711 letters come from the edition's provisional transcriptions, which the editors warn may still change before print. Sets marked "unchecked" have not had the editors' second reading.
- "Existing English version" records what has been checked so far. "Not yet checked" means only that. It does not mean no English version exists.
Rights and what is held back
Originals are quoted from the open-access texts of the Akademie-Ausgabe (Sämtliche Schriften und Briefe) published by the editing centres in Hannover, Potsdam and Berlin, and from public-domain printings.
Works the Akademie-Ausgabe has not reached, such as the Principes de la nature et de la grâce, are translated from Gerhardt's edition of 1875 to 1890, which is in the public domain.
Texts from Series II (philosophical correspondence) and Series VI (philosophical writings) are edited by the Leibniz-Forschungsstelle Münster, whose terms of use forbid publication, in whole or in part, without written permission. No text from those two series is reproduced here. Where a work in those series was also printed in an older public-domain edition, the translation here is made from that printing and says so. Otherwise the site holds the text back and shows only the reference until permission is on file.
The English translations and the commentary are the work of this project.