1. An Impossible Question

A father is fifty-six and his son twenty-nine. How many years must pass before the father is twice the son's age? If t denotes the requested number of years, 56 + t = 2(29 + t), so t = −2.

In 1831, Augustus De Morgan discussed this kind of result in a book about learning mathematics. He called the outcome absurd under the original interpretation, then used the sign to repair the question. The stated relationship held two years earlier, not two years later.

His treatment can sound extraordinary to a modern reader. Britain already had railways, steam printing, and educational publishing directed beyond a narrow learned elite. More than twelve centuries had passed since Brahmagupta set out signed arithmetic in 628, and the Chinese Nine Chapters preserved still earlier procedures involving positive and negative quantities.

The comparison readily becomes a morality play: the East accepted negative numbers while the West resisted them for a thousand years. Its simplicity is precisely what needs examination. What does accepted mean in each half of the sentence?

Notice first that De Morgan handles the negative sign accurately. He interprets it as information about the direction of time and diagnoses an overly restricted question. He is not unable to calculate or incapable of understanding what the result tells him.

His difficulty concerns a different claim: whether an isolated negative quantity can belong to arithmetic understood as operations on ordinary positive magnitudes. Subtracting something from nothing seemed no more directly conceivable in that framework than taking a square root of a negative quantity.

At least four questions must be separated. Can a negative term occur during an algorithm? Are there stable rules for calculating with such terms? Can a negative result be retained as the final answer to a particular kind of problem? Are negative numbers recognized on the same general footing as positive numbers?

These questions need not receive affirmative answers together. A practitioner can manipulate signed coefficients while accepting only positive roots. A debt can be a meaningful answer even when a negative length is rejected. Someone can defend negative numbers in one setting while retaining an incompatible explanation in another.

A misleading comparison takes the earliest evidence for one question in one tradition and compares it with a much later dispute about another question elsewhere. The dates are real, but the supposed delay measures different objects.

De Morgan's example exposes that error particularly well. The sign has mathematical and interpretive value even while he refuses the status modern arithmetic grants the number. His practice is more capable than a simple label of rejection suggests.

This essay follows those different forms of legitimacy through Chinese, Indian, Arabic-language, and European materials. It does not assume that each tradition asks the same philosophical questions or advances through one compulsory sequence.

Nor can the surviving evidence justify saying that only Europeans ever wondered what a number was. The later European debate is especially visible in the sources considered here. That is a statement about a documented discussion, not a monopoly on foundational thought.

The original age problem remains useful throughout. As a question about future years, it has no nonnegative answer. As a question about signed displacement in time, its answer is −2. The calculation, the domain of the question, and the status of the result are three distinguishable matters. Much of the history becomes clearer once they are no longer forced into one verdict.

2. Opposite Gains and Losses

The fangcheng chapter of the Nine Chapters introduces procedures for positive and negative rod entries. They arise in arrangements of quantities transformed by operations analogous to eliminating terms between columns. Subtraction can produce opposite states that must remain distinguishable in the subsequent calculation.

Liu Hui's third-century commentary explains their naming through opposed gain and loss. It also describes red rods for one class and black for the other, with an alternative distinction by the rods' arrangement when colors are unavailable.

The material point matters. A sign need not begin as a mark written beside a numeral. A rod's color, orientation, or position can encode the distinction directly within the working configuration. The evidence describes a practice rather than a modern integer printed with a minus sign.

As the preceding essay noted, textual instructions concerning red and black rods do not establish that every Han practitioner possessed and used a standardized two-colored set. The surviving objects and the prescription must be evaluated separately.

What the source strongly supports is the stable use of signed entries in a calculation. A negative entry can result when a positive entry is subtracted from an empty position. Rules for adding and subtracting like and unlike categories allow the procedure to continue.

What it less directly establishes is an explicit theory of −3 as an abstract object independent of every arrangement and application. That does not prove the author denied such an object. It marks the difference between the questions the text answers and those a modern reader may bring to it.

On the counting surface, an entry has a place, neighbors, and a role in an operation. Removing it conceptually from that setting and asking for a general ontology of signed integers is an additional philosophical move. Its absence from a procedural chapter should not be treated as a demonstrated inability to make it.

The early passage does not give a complete textbook of every operation on signed numbers. Its central work concerns addition and subtraction within fangcheng. Describing this as merely an error or a temporary failure would underestimate it; describing it as an explicit modern construction of the integers would add evidence we lack.

The source essay's metaphor of a verb becoming a noun captures something useful if kept metaphorical. An operation creates a state, and a name lets that state participate in further operations. But a relational name does not logically prevent the named item from functioning as a mathematical object.

Later Chinese work expands the range. Qin Jiushao's root-extraction procedures operate on arrays of polynomial coefficients with positive and negative statuses. Yang Hui reorganizes and explains earlier methods. Zhu Shijie's tianyuan and four-element techniques make signed coefficients part of sustained polynomial computation.

The development is therefore not confined to a single elimination trick. The role of the negative broadens within a computational tradition. Its legitimacy as a final answer still depends on the type of quantity and the problem's conditions.

Under the four-part distinction, intermediate use is early and explicit. Operational rules are available within specified procedures and later extend further. Acceptance of a final result requires contextual examination. A fully articulated account of signed integers as abstract objects is not supplied by these passages.

That last sentence is deliberately different from saying nobody in the tradition ever asked a foundational question. An archive of selected calculation texts cannot support a universal negative about centuries of thought.

The historical discipline is to avoid comparing one tradition's strongest computational evidence with another's most restrictive philosophical statement as though both answered one question. The rods show a successful practice. Understanding its success requires attention to what the rods, operations, and explanations actually do.

3. Property and Debt

Brahmagupta's Brahmasphutasiddhanta, composed in 628, gives explicit rules involving positive quantities, negative quantities, and zero. Its technical vocabulary draws directly on wealth or property, debt, and emptiness. These are not merely metaphors attached by modern teachers.

The rules cover the familiar behavior of signs. Two fortunes add to a fortune; two debts add to a debt. A fortune and a debt combine by taking their difference, with equal magnitudes yielding zero. Adding zero preserves the quantity. Subtraction changes the appropriate relationship between the opposed categories.

Multiplication and division receive corresponding sign rules: like signs yield positive results and unlike signs negative ones, subject to the separate problem of division by zero. Multiplication by zero yields zero. This is systematic signed arithmetic, not an isolated device for one numerical example.

The organization matters. General rules can be recalled and applied across problems. They make the operational distinction available as part of a body of mathematical knowledge rather than requiring every practitioner to rediscover it from a debt story.

Zero prevents the account from becoming a triumphal announcement that the complete modern system had arrived. Brahmagupta treats zero divided by zero as zero, and describes a nonzero quantity divided by zero in a form retaining zero as denominator. These prescriptions do not provide division as it is defined in modern fields.

It is more accurate to distinguish the successful sign rules from the unresolved or differently conceived treatment of zero division. The same author can be systematic in one domain and not supply a modernly acceptable extension in another.

Nor should uncertainty about division by zero be reduced to a simple failure to notice a problem. A surviving prescription belongs to a particular conceptual and textual setting. Later mathematicians would clarify why unrestricted division cannot coexist with the ordinary laws when the divisor is zero.

Five centuries later, Bhaskara II's algebra offers another complication. A passage commonly translated through Colebrooke rejects a second, negative value in the problem at hand and speaks of the nonacceptance of a negative absolute quantity.

The qualifier matters. A result intended to count objects or give an ordinary magnitude may fail the problem's conditions even in a tradition that calculates skillfully with debts and opposite directions. Rejecting that root is not the same as refusing all signed operations.

Selecting only Brahmagupta yields the slogan that India completely accepted negative numbers early. Selecting only Bhaskara's rejected value can yield the opposite slogan. Both flatten the distinction between operational legitimacy and admissibility as a particular answer.

A modern classroom still discards a negative root when a variable represents a nonnegative length or a count. What differs historically is the broader conceptual account surrounding that act. Modern students are usually told that the rejected value remains a number but is outside the problem's domain.

The historical passage does not automatically tell us everything Bhaskara thought about numberhood outside that example. It gives a judgment with a stated context, expressed through inherited categories of quantity. We should not turn either the acceptance or rejection into a complete creed.

Debt and wealth make opposition readily interpretable. So can east and west, increase and decrease, or a balance relative to a chosen reference. Five fewer monkeys than a reference population can be meaningful; negative five monkeys as the absolute size of a group is not. The formulation of the quantity matters as much as its name.

The age problem lies exactly on this boundary. Time displacement has two directions, but the original question explicitly asks for a future interval. De Morgan preserves the information in the negative sign by widening or reversing the question rather than treating the original wording as satisfied.

This is a mathematically intelligent response even though his 1831 account of arithmetic does not grant isolated negatives the status familiar today. The historical comparison becomes more revealing when each author is allowed to distinguish a usable operation, an interpretable quantity, and an acceptable answer.

4. Deficiency and Excess

Al-Khwarizmi's six forms, discussed in the preceding essay, keep the relevant squares, roots, and numerical terms positive. Restoration and cancellation transform expressions into those admissible forms. His book does not freely deploy signed coefficients in one universal quadratic equation.

That restriction belongs to a particular presentation, not to an undifferentiated Arabic prohibition. Later algebraic work widened the use of subtracted terms and signed coefficients substantially.

Al-Karaji's polynomial calculations allowed terms involved in subtraction to participate in extended operations. Al-Samaw'al explicitly stated sign rules and described the negative side through deficiency or diminution and the positive side through increase or excess. He could calculate a negative result from subtracting a positive term from an empty place, or the reverse when subtracting a negative term.

A coefficient embedded in a polynomial array was therefore not automatically excluded because of its sign. Operational scope could exceed the range of roots treated as solutions of a problem.

The vocabulary resembles Liu Hui's gains and losses in one respect: a pair of opposed situations supplies names for mathematical distinctions. That resemblance is worth noting without making it evidence of direct borrowing across the intervening geography and centuries.

It also does not prove that the entries were incapable of being objects. Modern mathematics routinely names objects relationally. The safer inference concerns the interpretation offered by the sources, not a psychological impossibility supposedly demonstrated by their etymology.

Positive roots remained natural in many problems because the root represented a length, amount, or count expected to be positive. A negative coefficient and a negative solution do different jobs. Their histories should not be measured by a single yes-or-no test.

Even the statement that money lacks a negative counterpart requires care. A physical pile of coins cannot have negative size, while a financial balance can represent debt. The variable's intended role determines which interpretation is relevant.

The sixfold classification thus displays a boundary in one algebraic framework. It does not prove that its author could not understand debt, and it does not fix the boundary for every subsequent Arabic-language mathematician.

Leonardo of Pisa's Liber abbaci offers a comparable pattern in a different setting. First composed in 1202, it survives through the revised tradition of 1228 rather than through an extant first-edition manuscript. Commercial and partnership problems can interpret a formally deficient amount as debt.

Such interpretations permit a useful result without requiring a general philosophical declaration that negatives are numbers identical in status to positives. The calculation and the practical meaning can work together before the broader category is settled.

But merchants should not be idealized as universally free of conceptual difficulty. Their work offered many naturally opposed quantities—receipts and payments, assets and liabilities, profit and loss. That made some negative outcomes easier to interpret. It did not make every quantity in commerce signed or every historical practitioner equally comfortable.

An operational success in one context is neither a complete ontology nor a meaningless trick. The historian's task is to identify the scope of that success. What transformations are legitimate? Which results are kept? What does the answer claim to represent?

Across the traditions considered so far, negative entries and sign rules appear earlier than a uniform modern explanation of all their roles. But the chronology remains uneven, and the examples do not amount to a worldwide inventory.

The contrast now shifts toward the especially visible European debate about number and magnitude. Its importance lies not in proving that one civilization asked uniquely serious questions but in showing how inherited definitions could conflict with increasingly successful symbolic operations. The difficulty could be argued, not merely felt.

5. What Is a Number?

Book VII of Euclid defines a number through a multitude of units. In that conceptual setting, three is three units and seven is seven units. Continuous magnitudes—lengths, areas, and volumes—belong to a related but distinct domain.

What would negative three count? What positive magnitude could be smaller than nothing? These are not foolish questions if the inherited definitions are kept unchanged while the algebraic operations expand.

Euclidean distinctions exercised substantial authority within several European educational traditions. They were not the unchallenged first lesson of every European for two thousand years, and scholars did debate the meanings of number, magnitude, and ratio. Nevertheless, their influence helps explain why formal success did not automatically settle conceptual status.

If subtraction means taking a quantity away from a quantity that contains it, a − b has that interpretation only when a is at least b. To make the expression meaningful more generally, something must change.

One route changes subtraction from a physical removal into a formal operation. Another interprets quantities through direction or opposed states. Another constructs a larger numerical domain in which each element has an additive inverse. Modern treatments can combine these approaches, but they should not be assumed present in every earlier use of a minus sign.

The resistance therefore had a logical content. Accepting a new result could require revising an old account of an operation or object that already supported much other reasoning. The problem was not merely finding a better picture of a debt.

That explanation must not become a compliment to European philosophical seriousness at everyone else's expense. Practice within Europe itself often moved faster than foundational agreement. Commercial arithmetic, curve theory, and coordinate methods did not wait for one universally approved definition of number.

Antoine Arnauld's proportional objection makes the conflict particularly clear. In modern notation, −1:1 and 1:−1 yield the same quotient, −1. Cross-multiplication also gives equal products. Yet if ratios are interpreted through ordinary comparisons of positive magnitudes, the first appears to place the smaller before the larger and the second the larger before the smaller.

The familiar inference that one ratio must be below unity and the other above it is not valid once the relevant signs are allowed. The apparent contradiction comes from extending symbolic rules while retaining an interpretation whose domain was positive magnitude.

The argument, commonly associated with Arnauld's 1667 geometry, is more serious than an expression of discomfort. It asks which inherited rules of proportional comparison survive an extension to signed quantities.

Leibniz treated the objection as worth answering rather than dismissing it because the symbols could be manipulated. Interpretations involving opposition, direction, or useful mathematical fictions offered ways to revise the relation between the formal operation and ordinary magnitude.

We should not turn that engagement into a fabricated quotation saying that Leibniz conceded everything to Arnauld. His writings and correspondence require precise dating and textual comparison. What matters here is the documented seriousness of the issue, not an invented decisive sentence.

The example identifies a recurring mathematical problem: two sets of individually useful conventions can become incompatible when used together outside their original domains. Resolution requires specifying which laws are retained and which interpretations are restricted or replaced.

That is the turning point in this essay. Earlier sections asked how negative entries were used. Here the question is what framework can justify the use without pretending that every property of positive magnitudes remains unchanged.

The negative did not become legitimate merely because someone finally pictured it vividly enough. Some of the work consisted in learning that an old picture had been granted authority over too large a domain.

6. Absurd, Fictitious, False

The historical labels cannot be collected as though they were interchangeable names for one fixed class. An author can call a problem absurd without first constructing negative numbers as a general category. A later author can call a result fictitious while allowing it to do useful work.

Diophantus's rejection of conditions that would produce negative solutions belongs to the setting of his problems. Michael Stifel's 1544 terminology more directly includes notions of absurd or fictitious numbers below zero. Cardano's Ars magna, published in 1545, similarly uses categories of fictitious quantities and equations.

Fictitious need not mean worthless. A quantity may be operationally productive while lacking an accepted ordinary-magnitude interpretation. The label can provide a provisional place within calculation rather than expel the object altogether.

Descartes's false roots in 1637 generally concern negative roots. His discussion of imaginary roots concerns another distinction, including roots that cannot be represented as real quantities. Conflating false, fictitious, impossible, and imaginary would erase the boundary between this essay and the next.

John Wallis supplies an especially revealing combination. His 1685 algebra explains negative quantities through movement: if a positive distance describes travel in one direction, a negative one can describe travel in the opposite direction. A line extending both ways provides an intelligible representation.

Yet Wallis also explored arguments in which negative ratios appeared to lie beyond infinity. As a positive denominator decreases toward zero, a quotient grows without bound; treating passage through zero as a continuation into another region encouraged an interpretation incompatible with the ordinary ordered real line.

These discussions belong to different contexts within Wallis's work and should not automatically be compressed into one page or one moment. Their coexistence within his mathematical career is the important point. A powerful geometrical interpretation did not instantly clear away every older proportional intuition.

The history is not well described by two teams, acceptors and rejecters. Stifel could use what he called absurd. Cardano could calculate with what he called fictitious. Descartes could transform roots he called false. A label identifies a status within a practice, not necessarily a refusal to operate.

Newton used affirmative and negative quantities in arithmetic and geometry and investigated curves across the plane. His classification of cubics appeared in 1704, not first in 1710. Whatever his presentation lacked by later structural standards, his mathematical practice made extensive use of signed relations.

Euler's algebra, published around 1770, explains signs through property and debt, acquisition and loss, and quantities above and below zero. Operational hesitation is much less prominent than in De Morgan's early discussion. Yet several interpretations still stand together: subtraction as an operation, negativity as a property, and sign as direction or financial state.

It would be misleading to say Euler did not understand negatives. His practice was highly developed. The point is that a modern unified structural account should not be projected into every explanatory phrase he used.

De Morgan's own position changed. His 1835 engagement with Peacock expressed unease about symbols apparently seeking meanings. By the 1840 article on negative and impossible quantities, he articulated a sequence in which symbols and their laws are specified and consequences developed. His 1842 work on the foundations of algebra pushed further toward explicit structure.

That development did not make meaning irrelevant to him. He continued to value interpretations and expected successful symbolic systems to acquire broader ones. The change concerned when an interpretation had to be available and what its absence prevented.

This is why the father-and-son example should not become a lifelong epitaph. A quotation from 1831 identifies a position at a date. It cannot stand for every later thought of a mathematician whose work helped alter the terms of the discussion.

The negative vocabulary records tension, but it also records accommodation. Naming a thing fictitious could be a way to keep it in play while its legitimacy was being renegotiated. The next essay will show that provisional status doing even more conspicuous work.

7. Let It Work First

A recurring pattern crosses the geographical divisions. Signed terms can function within polynomial operations while negative roots remain restricted. A fictitious answer can guide further calculation. A curve can be drawn using quantities whose philosophical interpretation remains unsettled.

De Morgan explicitly recognized a version of that historical pattern: the ability to produce results can precede the ability to explain them satisfactorily. Repeatedly successful operations may deserve retention while a more general interpretation is sought.

This does not amount to a rule that any procedure producing a few correct answers should be trusted. The question is how a developing practice earns enough stability to justify further investigation before its final account is available.

Earlier essays supplied related but distinct boundaries. Archimedes distinguished mechanical discovery from the demonstration he required. Liu Hui left a volume problem unresolved rather than claim a reason he did not possess. De Morgan reflected on a broader history in which explanatory frameworks were widened after effective procedures had already appeared.

The comparison is interpretive. These authors did not share one theory of progress or one standard of proof. What connects them is a visible difference between obtaining a result and explaining the grounds on which it may be accepted.

One important nineteenth-century route was symbolic algebra. George Peacock's 1830 treatise distinguished arithmetical algebra, whose symbols retained the restrictions of ordinary arithmetic quantities, from a more general symbolic treatment governed by stated operational laws.

The principle later described as permanence of equivalent forms sought to preserve relations valid under legitimate arithmetic operations when the symbols received an extended interpretation. It was not a magical warrant for every conceivable extension. The retained laws and their compatibility still required attention.

The change lowered an entry barrier. A symbol did not first have to identify an ordinary positive magnitude before it could enter a legitimate algebraic investigation. Its relations under specified operations could become the starting point.

Hermann Hankel's 1867 work gave an influential formulation of the permanence of formal laws in the extension of number systems. New objects could be investigated through a system of operations rather than always reduced to an old kind of magnitude.

This was not yet a declaration that consistency questions or interpretation could be ignored. Nor could every law be retained in every extension: later examples make clear that enlarging a system sometimes requires sacrificing a familiar property. The historical principle guided construction; it did not remove the need to verify it.

Geometry supplied another route. Directed segments made −3 interpretable as three units in the opposite direction rather than a positive length mysteriously smaller than nothing. Wallis's diagram and Newton's uses of opposite directions contributed to that practice. Descartes's analytic geometry helped enormously without immediately settling every question about negative roots.

The representation answers an important intuitive problem, but it is not the entire logical construction of the integers. Debts, temperatures, and abstract roots are not literally segments. A picture alone does not establish every algebraic law.

Applications supply the third route. Repeated use in commercial balances, motion, coordinates, and equations stabilizes meanings and makes signs ordinary. It also supplies tests that can expose inconsistencies in a proposed explanation.

The relative importance of these routes depends on the question. For a general algebraic status independent of an immediate magnitude interpretation, symbolic approaches are central. For why practitioners and students came to find negatives intelligible, geometry and applications cannot be reduced to ornamental illustrations.

It would be too neat to call the outcome formalism defeating intuition. Much nineteenth-century work negotiated among extension, rigor, and intelligibility. The desire for wider operations did not eliminate the desire for reasons or representations.

The essay's philosophical suggestion is that legitimacy sometimes changes when an admission requirement changes. Negative numbers did not need to become secretly positive magnitudes. A larger system could be built in which they had defined relations and operations.

That is different from saying numbers stopped referring to anything concrete. They retained many interpretations and acquired others. What changed was the insistence that one particular interpretation must precede every lawful operation.

8. What the Names Remember

Negative derives through a vocabulary of denial or negation. Positive comes from a semantic field of placing, affirming, or setting down. Minus means less. These words did not begin as neutral labels invented for a modern abstract integer system.

Comparable relational vocabularies occur elsewhere: Sanskrit debt opposed to fortune, Arabic diminution opposed to increase, Chinese loss opposed to gain in Liu Hui's explanation. The names preserve older settings in which opposition became computationally useful.

That is the basis of the title's image of a wound in the name. A later object can gain equal formal standing while its inherited label continues to suggest deficiency or exclusion. The metaphor should not be mistaken for a universal linguistic law or proof that the object was never independently conceived.

Chinese fu has neighboring senses involving carrying, owing, loss, and failure. Which etymological path directly produced its mathematical use cannot be recovered simply by choosing the most appealing modern gloss. Liu Hui's own explanation of opposed gain and loss is firmer evidence than a speculative origin story.

Nor should fu automatically be translated as debt in every early Chinese context. That would make the Chinese source look more like the Sanskrit tradition than its own wording warrants.

Late Qing translators did not need to invent positive and negative from nothing. A long-established mathematical vocabulary was available. Li Shanlan and related mathematical texts could use zheng and fu, including explicit sign rules, while reorganizing them within imported forms of algebra.

The continuity of the words does not imply that nothing changed. A distinction used in rod arrays, then polynomial coefficients, could be assigned a new role in a nineteenth-century account of signed numbers. An old term could bridge systems without having traveled along the same route as the European concept it translated.

The preceding essay made a similar point about suanfa and algorithm. Translation can align two independently developed vocabularies. A familiar Chinese expression need not be a phonetic or semantic descendant of the European word it now routinely renders.

Where evidence for competing Chinese translations or the earliest formation of later pedagogical terms is lacking, the gap should remain visible. A plausible invented first user closes inquiry prematurely. An explicit uncertainty leaves a question another scholar can investigate.

Return to the four senses of acceptance. Negative entries appear early in Chinese procedures and elsewhere in computational traditions. General sign rules are explicitly developed in Brahmagupta and later in Arabic-language algebra, with their own histories in Europe.

Final answers remain conditional everywhere because a variable's intended domain matters. A signed balance is not the same kind of quantity as the absolute number of animals in a group. This is not a historical embarrassment; modern mathematics also distinguishes an equation's formal solutions from the admissible solutions of a modeled problem.

The fourth question concerns general numerical status. The European debates examined here make that question especially explicit and connect its answers to changes in symbolic algebra, geometry, and definitions of operations. They do not prove that other traditions never considered foundational questions or that one debate alone created legitimacy for the whole world.

The century-spanning contrast at the opening must therefore be rewritten. It is not one clock measuring how long different peoples took to understand a simple idea. It is several histories of operations, applications, admissible solutions, and explicit foundations.

De Morgan's change of position is valuable because it shows those histories meeting inside one career. A mathematician can use a negative sign correctly, reject one interpretation of it, and later help formulate a broader account under which it acquires another status.

The negative did not win recognition through a single metaphor, a single rule, or one official decision. It accumulated stable operations and useful meanings while inherited restrictions were reconsidered.

Another kind of quantity will sharpen the pattern. Its name still suggests that it is imagined, and it entered crucial calculations even when no debt or reversed direction seemed ready to interpret it. The problem was not merely how to tell a better story about the object. Sometimes refusing the object meant losing a result that the existing methods were supposed to deliver.