Insurance by BogdanovaThe Encyclopedia of Insurance and Risk Science

RESEARCH DOSSIER

Sea Loans, Usury, and the Origins of Insurance Contracts

Abstract. The differentiation of maritime credit from marine insurance is a central episode in the history of contracts, financial intermediation, and commercial institutions. Ancient contingent sea loans placed maritime casualty risk on lenders whose principal and return depended on a voyage's successful outcome. Medieval commercial innovations gradually detached that protection from the advance of funds, allowing merchants to pay separately for a contractual promise of compensation. This monograph examines the documentary and conceptual boundaries among ancient trade-risk allocation, Greek maritime loans, Roman pecunia traiecticia, medieval sea credit, ecclesiastical prohibitions of usury, Genoese fictitious loans and conditional sales, and explicitly priced marine insurance. It compares ancient legal texts with modern archival studies of Genoa, Palermo, Florence, Prato, Barcelona, and Venice, emphasizing the difference between a contract's stated legal form and its actual economic performance. Particular attention is devoted to the evidence for Genoese and Palermo instruments in 1343, the Palermo deed of 1347, late-fourteenth-century Datini insurance, and the contrasting legal cultures of Genoa and Florence. The central historical conclusion is that the emergence of insurance as a distinct bargain depended on separating protection from credit while making the insured subject, covered contingency, obligated risk bearer, price, and settlement procedure independently recognizable. Religious regulation influenced contractual form, but changes in merchant liquidity, transport networks, adjudication, and risk-bearing capacity also shaped the process. The work offers an original comparative interpretation of established evidence; it does not claim to establish a new doctrine, a single universal origin, or previously unknown archival discoveries.

Keywords: marine insurance; sea loans; bottomry; respondentia; foenus nauticum; pecunia traiecticia; usury; medieval canon law; Naviganti; commercial contract history; Genoa; Palermo; Florence; Francesco Datini; conditional sales; insurance premium; maritime finance; source criticism; risk transfer; history of insurance.

Introduction

Marine commerce required promises whose value depended on uncertain events. A cargo might arrive intact, be captured, be thrown overboard, suffer deterioration, or disappear with the vessel. Merchants could finance voyages from their own wealth, borrow from lenders, organize partnerships, transfer some exposures through agreements with other merchants, or distribute loss among participants in a commercial enterprise. These responses developed in different places and periods. Their shared reference to danger does not establish an identity of legal form.

A maritime loan makes an advance of capital whose repayment is conditioned upon maritime circumstances. A marine insurance agreement grants protection against a defined loss in return for consideration, without requiring the insurer to have advanced the funds being protected. This distinction is the analytical center of the present study. A sea loan can perform part of the economic work of insurance, yet its obligee, initial cash flow, and contingency structure remain those of a credit contract. Conversely, an early insurance promise can appear in an archive under the name of a loan or sale even when no genuine lending or purchase occurred. Contract names therefore cannot settle institutional identity.

Historians have long disagreed about the degree of continuity between ancient sea loans and medieval premium insurance. De Roover's study of early insurance documents treated the ambiguities of the surviving sources as a fundamental problem. Recent large-scale archival research by Botticini, Buri, and Marinacci (2023) has substantially improved our understanding of the medieval markets, their participants, and the risks represented in the records, while also preserving uncertainty over a single point of invention. Their study of insurance contracts and markets includes thousands of contracts and legal disputes from several cities rather than resting its historical interpretation on one ceremonial first policy.

The research follows a chronological and contractual inquiry. It first identifies ancient contingent obligations, then the shifting legal treatment of risky credit, then medieval transformations of the contractual object, and finally the appearance of observable prices and specialized risk-bearing markets. Throughout, the question is what a party was legally required to do upon safe arrival, loss, delay, partial salvage, or deviation. The analysis does not assume that the earliest surviving document marks the first occurrence of the practice, and it distinguishes documents read directly in available editions from notarial originals known here through published archival transcriptions and scholarly reconstructions.

Research Object, Sources, and Method

The historical object is the transition from combined finance-and-risk agreements to independently purchased protection over voyage-related losses. Three kinds of evidence are brought into dialogue. First, published primary legal texts state the operative rules or present actual contractual language. Second, later scholars' editions and archival studies reconstruct individual medieval deeds whose physical originals are kept in repositories not directly inspected during this research. Third, economic historians use larger samples to compare market participants, pricing, routes, and institutional change. Each evidentiary tier answers a different question.

The Code of Hammurabi, the Athenian speech Against Lacritus, the Roman Digest on maritime interest, and the canon-law decretal Naviganti are textual witnesses to rules and disputes. None is a comprehensive ledger of how all merchants behaved. A law may prohibit a transaction that nevertheless occurs; a litigant may describe its opponent strategically; a transmitted decree may be assigned different dates in a later secondary chronology. These problems require close attention to genre, language, jurisdiction, and manuscript transmission.

For medieval practice, Balard's (1978) Genoese documentary analysis, de Roover's (1945) comparison of early insurance documents, and Iodice's (2023) research on Genoese underwriters provide complementary views. Botticini, Buri, and Marinacci (2023) assemble evidence from contracts and court files to show how techniques and prices varied across cities. The present work treats published summaries of individual deeds as mediated archival evidence: the deed is a historical primary document, but its reading here passes through the cited investigators. This prevents the unwarranted claim of having examined original notarial folios.

Five checks structure the contract comparison. One asks whether a genuine advance of money was made. Another identifies who owes a payment under each eventuality. A third examines whether the object is a debt, a purchase, or an independently indemnifiable interest. A fourth asks how consideration is expressed and whether a risk-specific premium is observable. A fifth considers how voyage definition, claims notice, proof, and judicial or notarial enforcement shaped performance. These are ordinary comparative questions in legal and economic history, not a proposed new doctrine. Together they provide a transparent way to compare sources whose vocabulary and legal settings vary.

The boundaries are equally important. The surviving material heavily represents Italian and Catalan commercial centers, certain well-documented merchant families, and jurisdictions with recoverable notarial or account-book archives. It does not justify a statement that no comparable protections existed anywhere else. The formal differentiation of marine insurance does not date the emergence of every earlier system of mutual aid, loss sharing, or public relief. Nor does a medieval premium record establish modern actuarial calculation; pricing could be experiential, negotiated, or supported by trade information without a fully specified probability model.

Contingent Marine Loans in Ancient Practice

Ancient commercial law already distinguished ordinary loss in an agent's undertaking from some losses outside the agent's control. The Babylonian Hammurabi provisions 102–105 concern merchants, agents, accounts, and travel-related loss. Provision 102 assigns ordinary business loss to an agent in the cited English translation, while provision 103 releases an agent whose goods are taken by enemies on a journey after a prescribed oath. This evidence shows a legally recognized allocation of involuntary commercial loss. It does not contain a premium, an insurer's compensatory promise, or a named policy of marine insurance.

The distinction matters because histories of insurance frequently classify every early allocation of misfortune as proto-insurance. Such a classification may illuminate a broad practical concern, but it can also obscure the institutional characteristics that require explanation. A statute exonerating an agent from liability when robbed shifts the burden of loss within an existing merchant-agent relationship. There is no indication that an independent third party sold a priced undertaking to reimburse the merchant. The safest inference is therefore that ancient legal systems addressed who should bear involuntary commercial losses, while the presence of an insurance market must be demonstrated by additional evidence.

Long-distance trade in the Mediterranean later generated arrangements in which merchants placed money at risk with the voyage. Ancient Greek maritime lending is comparatively well documented because Athenian court speeches reproduced or summarized contractual terms. These cases concern the financing of cargo purchase and carriage. The lender's expectation of a higher repayment when the voyage succeeded was inseparable from exposure to the voyage's possible failure. The lender could diversify by financing several expeditions, but the surviving contractual language shows the lender as financier and risk bearer within the same transaction.

A contract's conditional nature does not make every payment contingent on every imaginable danger. Particular claims could depend on the ship's route, the survival of pledged goods, deviations authorized by agreement, agreed jettison, or the period during which the lender carried the risk. A careful reconstruction must recover these boundaries rather than write a generic formula that debt vanished whenever the vessel suffered any misfortune. The technical detail of ancient voyage finance already points toward later insurance practice: contract construction organizes risk by subject, peril, time, and compliance.

Greek Maritime Credit and the Documentary Contract

The speech commonly attributed to Demosthenes, Against Lacritus, sections 9–13, offers an unusually revealing passage: the plaintiff presents a written maritime loan agreement in a dispute over performance. According to its transmitted English translation, Androcles and Nausicrates lent 3,000 silver drachmae to Artemon and Apollodorus for a voyage from Athens through Mende or Scione toward the Black Sea and back. The pledged security included wine transported on a named vessel. This was a commercial loan embodied in a specific documentary arrangement, not a bare promise to pay if the borrower became unfortunate.

The agreement differentiated a return of 225 drachmae per thousand from 300 per thousand if departure from the Black Sea occurred after the rising of Arcturus. These figures correspond to voyage-contingent returns of 22.5% and 30% on the stated principal; they should not be presented as annual interest rates. The difference reflects the contracted itinerary and seasonal timing as recorded in the speech. Its interpretation as compensation for changing risk is economically plausible, though an individual contract does not isolate the exact share attributable to storm hazards, the time value of money, or bargaining conditions.

The security and repayment clauses prohibited further loans on the pledged goods, required the return cargo to accompany the vessel, and made payment due after safe arrival at Athens. A further term allowed lenders to seize or sell collateral and pursue other assets following nonpayment. The document also addressed jettison and money paid to enemies. These terms reveal an intricate allocation: lenders accepted specified maritime exposure while retaining strong remedies against deliberate nonperformance or ordinary default after the contractual condition was met.

The litigation illustrates another crucial distinction. The allocation of accidental loss is separate from the allocation of misconduct. A borrower could violate route or collateral requirements, misuse funds, or evade the creditor notwithstanding a conditional loan. An agreement that releases the borrower when covered perils destroy the venture does not license the borrower to fabricate a loss or defeat the lender's contractual remedies. Such distinctions help explain why detailed voyage clauses were valuable to both financiers and later insurers.

Another speech, Against Dionysodorus, attests to the broader Athenian legal world of maritime borrowing and disputes over routes and repayment. It is used here principally as corroboration of a contract-centered maritime-credit culture, rather than as a source of unverified figures or a reconstruction of every paragraph. The Athenian speeches are forensic advocacy; their detailed contract passages have greater evidentiary weight for agreed terms than the speakers' accusations about the moral character of their opponents.

The economic significance is straightforward. Maritime lenders bore a casualty component of the venture's financing risk; borrowers paid a return that could be high relative to ordinary credit because expected repayment was linked to the journey. The lender's business combined provision of capital and assumption of hazard. A separate insurance policy would make it possible for the same merchant to obtain financing from one counterparty and loss protection from another. Ancient Greek sources explain the bundled starting point and the contractual skills upon which later distinctions could build, without demonstrating a continuous institutional chain across the intervening centuries.

Roman Maritime Loans: Risk and Interest

Roman juristic texts treat maritime lending as a distinctive legal situation. In Digest 22.2.1, Modestinus distinguishes money transported across the sea from money consumed locally, and asks whether merchandise purchased with the proceeds sails at the creditor's risk. The relevant condition is therefore not a maritime label alone: the transaction's designation relates to actual exposure during the voyage. In Digest 22.2.3, the creditor's risk begins from the date agreed for the vessel to sail.

Papinian's reasoning in Digest 22.2.4 connects the exceptional remuneration of maritime lending to the continued existence of the creditor's risk. Where the lender never bore the maritime risk or that risk had ceased under the agreed condition, extraordinary maritime interest was unavailable under the described rule. The text supports a historically important legal proposition: a higher return was treated as related to an actual hazardous contingency rather than simply to money carrying a maritime name.

These passages preserve distinctions between principal, contractual risk, interest, and security. The maritime loan's legal identity depended on the exposure of the creditor, while the lawful price of that obligation was conditioned by the timing of that exposure. This was not yet a stand-alone insurer's promise. The capital was provided as part of the transaction, and the anticipated payment to the lender arose through successful performance of the voyage. Roman law nevertheless supplied a rich repertoire for later commercial lawyers who needed to express contingent obligations with familiar juristic language.

A common historiographic shorthand calls bottomry a direct predecessor of insurance. The expression is serviceable only if the continuity is stated precisely. Maritime credit and marine insurance both allocate losses arising from navigation; they can differ in the direction of their payment, the presence of an actual advance, and the legal interest being protected. A merchant indebted on a sea loan may be relieved from repayment after a covered maritime casualty. A merchant buying insurance may receive a positive claim against an underwriter after the casualty. Those are economically related protections but legally different performances.

The Roman evidence is also a reminder that terminology changed. The expression foenus nauticum identifies maritime interest; pecunia traiecticia concerns funds placed at maritime risk. Later English-language terms such as bottomry and respondentia have narrower technical uses in some periods and broader retrospective meanings in others. Where a source is concerned with collateral in the ship or goods, that distinction should be explained rather than silently imposed on every ancient and medieval document.

Credit, Usury, and Commercial Norms

The medieval European emergence of separately contracted marine insurance cannot be explained solely by saying that lending became illegal. Moneylending, profit-sharing partnerships, commercial exchanges, bills, sea loans, and risk-bearing promises coexisted under different systems of religious and secular regulation. Their relative convenience depended on a merchant's available capital, the length and purpose of a journey, creditor remedies, expected transport hazards, and the costs of writing and enforcing a contract. The classification of a payment as interest, profit, compensation for risk, or sale price could be legally consequential, yet commercial actors had economic reasons to seek protection even apart from theological controversies.

Christian prohibitions of usury created particularly powerful pressures on the legal presentation of credit and other contingent arrangements. Scholarly histories describe maritime loans as initially receiving special treatment because the creditor bore the risk of the sea. A lender who could lose the principal in a shipwreck had a claim to remuneration different in character from an ordinary sure-return creditor. Yet the argument from risk did not remain an automatic defense of every excess payment. The canonical material would sharpen the distinction between risk exposure and permissible profit.

The language of moral economy and the terminology of commercial records need careful handling. Usura could function as a juristic and theological classification; an interest rate observed in a maritime deed is a historical economic datum. Equating every high rate with a legally established violation assumes the outcome of the very debate being investigated. Likewise, an agreement calling itself a gratis loan may in practice conceal compensation. Balard's discussion of Genoese contractual forms shows merchants using formulaic statements of gratuitousness in documents whose economic functions require separate reconstruction.

This interaction between commercial practice and legal categories had a substantial documentary consequence. A lender and a merchant needed language that a notary and court would accept. If a transaction moved money and contingency between the parties but the chosen written form could be challenged as usurious, the same commercial protection could be restated through exchange, fictitious loan, sale, or an overt insurance bargain. Some such formal changes could conceal fundamentally different economic performances; others may have reorganized only the enforcement pathway. Source criticism must decide case by case.

Merchant liquidity supplied an independent reason for change. Botticini, Buri, and Marinacci (2023) observe that wealthy Italian merchants increasingly could finance their own shipments and therefore had less reason to purchase an expensive bundle of external capital and casualty protection. Their historical explanation of separated protection emphasizes the interaction of this supply of self-financing with the demand for protection during long-distance trade. A merchant who already possessed working capital might still wish to prevent a single wreck or seizure from destroying an enterprise. Insurance could address that need without obliging the merchant to borrow.

The shift can be expressed with a simple transaction comparison. Suppose a merchant requires capital of 1,000 monetary units for cargo. Under a conventional sea loan, a lender advances the 1,000 and bears specified casualty risk; if the voyage succeeds the borrower returns principal plus the agreed maritime charge. A self-financing merchant requires no advance but may still wish to protect an exposure worth 1,000. Under separate insurance the merchant retains or raises capital through another means, pays a premium, and obtains a promised compensatory payment if the covered loss occurs. These are illustrative cash flows, not historical quotations from an individual deed, and they introduce no proposed pricing model.

The causal interpretation consequently has several parts. Canon-law restrictions shaped the legitimacy and enforceability of contracts. Growing accumulations of commercial wealth altered the demand for loan capital. Notaries, merchants, brokers, and courts supplied the infrastructure through which an independent risk promise could be expressed. Voyages and conflicts produced losses sufficiently concentrated on individual ventures to make the promise valuable. A single-factor explanation misses the interaction among these established features.

The Decretal Naviganti and Its Interpretation

The text known as Naviganti occupies a prominent place in the history of the sea loan. Its operative wording appears in the collection of Gregory IX's decretals under Liber Extra 5.19.19. The text characterizes as usurious a creditor who lends a definite amount to someone sailing or traveling to fairs, assumes the risk, and expects to receive more than principal. This precise juxtaposition is significant: assumption of risk, by itself, did not prevent the theological-legal condemnation described in the decretal.

The source is regularly described in later economic history as the condemnation of maritime loans in 1236. The published canonical collection is identified as the Decretals of Gregory IX of 1234, and the online source edition provides the precise location X 5.19.19. These chronological labels refer to different documentary contexts and secondary traditions. The present research uses the stable canonical book-title-chapter reference for the text and treats the frequently cited 1236 designation as a secondary historical dating convention rather than proof that a new rule was first promulgated at that moment. This distinction is essential when a purported exact date becomes the causal anchor for centuries of subsequent commercial practice.

The second half of the decretal's Latin text concerns certain forward transactions in goods and the uncertainty of later prices. Its juxtaposition with the loan language shows that canon-law reasoning attended to the substance and uncertainty of contractual gain. It would be misleading to read the first sentence as a complete ban on every commercial arrangement containing a maritime contingency, or the later discussion as a blanket approval of every disguised debt. Individual transactions required their own legal classification.

Historians have debated how rapidly the rule changed practice. A normative document can affect incentives by exposing merchants to the risk of a usury claim, altering the formulations available to notaries, or changing expectations about enforcement. But a prohibition's existence is not evidence that all sea loans ceased on publication. Evidence for continuing or revived bottomry in later centuries makes any claim of permanent disappearance untenable. Zanini's study of later Genoese maritime finance illustrates the continued use of conditional maritime financing, in a different setting and period.

One historically defensible inference is that the decretal increased the importance of form and substance in the documentary presentation of voyage risk. It helps explain why a transaction that was economically an insurance guarantee might be written as a familiar loan or sale with deliberately inverted conditions. But the text does not prove that Genoese insurance was invented by a papal decree, that the first hidden policy followed immediately, or that all maritime insurance claims thereafter turned on canon law. Those stronger causal claims require additional documentary chains and counterfactual evidence.

The distinction between a rule and its enforcement can be made concrete. A notarial record may contain a formal repayment promise capable of use in court; the economic bargain may include a separately transmitted premium absent from the instrument. The omission may have helped avoid a usury characterization, or it may reflect conventions of recording prices elsewhere. A historical interpretation should privilege contemporary explanations and the structure of the obligation over automatic attribution of motive. Balard's accounts of Genoese forms show the importance of unrecorded insurance consideration, while de Roover (1945) warns that surviving language can support competing interpretations.

Fictitious Loans and Conditional Sales

The decisive change in the legal-economic content of a purported loan occurs when a transaction is named a loan but its stipulated debt runs from the prospective risk bearer to the exposed merchant, and the supposed principal was never genuinely advanced. In a familiar sea loan, the merchant receives capital, and the lender has a conditional claim to repayment if the voyage succeeds. In the early Genoese fictitious loan described in the historical studies, the insured merchant is represented as a lender to the insurer, although no actual loan of the nominal capital has been made. The supposed debt becomes payable if the ship or cargo is lost and is extinguished if it arrives safely.

This reversal of conditional performance marks the insurance substance of the arrangement. The insured merchant is no longer primarily obtaining financed capital; the merchant is acquiring an enforceable claim against another person upon a defined adverse event. The insurer's promise is the financial response to the loss. Botticini, Buri, and Marinacci (2023) describe precisely this inverted fictitious-loan structure. The fact that the deed may state that a fictitious sum was lent gratis does not establish gratuitous protection: a premium might have been paid but omitted from the recorded instrument.

The difference in promised performance can be summarized without imposing modern legal labels on the notaries. In the ordinary contingent sea loan, a successful voyage produces a payment to the financier. In the fictitious insurance loan, an unsuccessful voyage produces a payment to the exposed merchant. The direction of the conditional claim and the absence of a genuine initial advance are more decisive than the word mutuum in the document.

The next prominent Genoese device was a conditional or fictitious sale. Under the reconstructed form, the person taking the maritime risk purported to purchase specified cargo or a vessel at an agreed price if it failed to arrive; the sale became ineffective on safe arrival. In commercial substance the purported purchaser promised to pay for a loss, while the nominal sale generally involved no actual transfer of the property before the contingency. Balard (1978) dates this form's appearance in Genoa to around 1362 and explains how the conditional-sale documentary device could be used to express an insurer's obligation within a familiar legal form.

An early sale disguise raises difficult interpretive questions. A sale can genuinely transfer ownership or future economic risk; a purported sale may leave the seller in possession, with no ordinary bargain to exchange goods. It would be inaccurate to label every conditional sale an insurance policy. The historian must establish whether the supposed purchaser was in fact a risk bearer who received a risk price, whether the contract was canceled on safe arrival, and whether failure of the shipment produced the promised payment. In these instances the legal words become evidence to be interpreted, rather than a final legal classification.

The historical literature places these forms in a larger sequence but not a universally simultaneous one. The Genoese notarial environment continued to use disguised structures after merchants elsewhere recorded more explicit insurance agreements. The question is thus not how a whole civilization advanced through an identical list of legal stages, but how different local commercial systems came to identify and price an indemnificatory promise. Explicitness in the deed, commercial separation from credit, and maturity of a market need not arrive on the same day.

A fictitious debt also helps explain enforcement incentives. A promise cast as an established debt might allow a claimant to use recognized procedures for recovery. Yet it could create its own vulnerabilities: the debtor might deny receiving the fictitious sum, plead illegality, challenge the contingency, or dispute the loss. The evolution toward expressly described insurance could reduce reliance on mutually understood fictions and make the object of the bargain more transparent. This proposition is an interpretation of contractual incentives, not a claim that every medieval legal forum applied the same remedies.

Genoa, Florence, and Distinct Contractual Cultures

Genoa and Florence demonstrate that a common commercial demand can generate different documentary institutions. Genoese marine insurance was closely associated with notarial deeds written in Latin and formulas that concealed the risk bargain under the apparent shape of a loan or sale. In Florence and its commercial network, written private agreements in the vernacular and the work of brokers could facilitate the negotiation of explicit cover without the same notarial presentation. The contrast emerges from the comparative archival findings of Botticini, Buri, and Marinacci (2023).

The difference should be interpreted as one of institutional practice, rather than a difference in what qualified as a loss. Both communities faced shipwreck, capture, deviation, delays, and commercial uncertainty; both developed sophisticated methods of allocating them. The Genoese document could specify perils with great precision while omitting the visible premium. A Florentine account book could reveal the premium and underwriters clearly while relying on trade conventions and broker records for other details. The historian who studies only the face of notarial deeds may mistake documentary silence for the absence of a price.

At Genoa, notaries provided standardized records and a degree of legal solemnity. Separate insurers could be bound through separate instruments. The formula preserved recognizable rights and conditions even when the real economic promise was hidden by a legal fiction. Balard's corpus of Genoese insurance deeds includes 1,169 acts dated from 1343 to 1400, predominantly from the later decades of that century. This is a scholarly assembled corpus, not a claim that those acts comprise every transaction completed in the city. The records nevertheless establish continuing use of the instruments and growing documentary density.

Florentine and Pisan merchants developed a different record culture. Account books, correspondence, and private instruments associated with the Datini network show cover being purchased and written as insurance. An official archival guide to the Datini collection reports surviving policies associated with Pisa and Florence and dates their earliest extant examples in that collection to 1379 and 1385, respectively. These are repository-specific survival dates, not absolute dates of invention in either city. The guide is used for its archival inventory and dates, not as independent proof of an earlier universal priority.

In the later expansion of insurance markets, the growing role of brokers connected an insured merchant with several possible underwriters. Brokers could help transform a local relationship between two merchants into a repeatable exchange in which terms and shares were negotiated among participants. Private contracting also enabled separate quotas to be placed under a common commercial instrument. The arrangement mattered because one large cargo loss need not rest on the fortunes of a single person.

Genoa and Florence also stand for a more general feature of institutional history: legal form may remain conservative while business function changes quickly. A deed may keep familiar language for judicial acceptance, even as the transaction underlying it has been transformed. Conversely, an explicit title may precede a complete regulatory and accounting system. Therefore the chronology of insurance cannot be reconstructed solely from the first use of the word insurance. The economic obligation, available evidence of a premium, and operation of the market must be considered together.

The Evidence of 1343 and 1347

The date associated with an early insurance document is a claim about surviving evidence and its classification. It is not proof of the day on which merchants first imagined transferring a maritime loss. Modern research identifies a deed drawn up at Genoa by the notary Tommaso Casanova on March 18, 1343 as a frequently cited early instance. Iodice (2023) refers to that Genoese notarial instrument of 1343, while Botticini, Buri, and Marinacci (2023) identify it as a document considered the first modern insurance contract in a conventional historical chronology.

The same newer research reports a 1343 insurance contract recorded at Palermo, involving a Genoese insurer and a Pisan insured merchant. The scholars state that the earliest contract in their dataset was drafted in Palermo in 1343. These pieces of evidence complicate the habit of presenting a single city and a single deed as an unquestionable moment of creation. The secure conclusion is that recognizable contracts were documented in more than one connected Italian and Sicilian commercial setting in the early 1340s. Whether all extant documents meet a single juristic definition of insurance is itself part of the historical inquiry.

An unusually concrete case is a deed drafted by the notary Stefano Amato at Palermo on February 25, 1347. Botticini, Buri, and Marinacci publish a structured description of the Palermo insurance of 1347. Francesco di Canigliano, a merchant from Pisa, was the insured; Baldassarre Grillo, a Genoese merchant, was the insurer. The goods, described as quantities of cheese and leather, were to travel by galley from Palermo to Pisa. The insured value was 100 gold florins and the recorded premium was 7.5%.

This instrument is significant for more than its age. It identifies distinct parties, a described interest, the route, a ship, a sum protected, an explicit premium, covered natural and human causes, and settlement procedures. The reported terms extended the guarantee to loading and unloading, permitted justified diversions, and specified a deadline to liquidate damage after reliable notice. The contract reveals the practical components of a separately purchased protection bargain, rather than a mere analogy between maritime danger and insurance.

A further Datini contract, dated December 19, 1384, involved shipment of valuable veils from Porto Pisano to Barcelona with cover of 130 florins and an insurance premium of 6%. The Datini policy and its loss report name separate underwriters taking quotas of 50 and 80 florins and report seizure by hostile galleys, followed by notification and a specified period for payment. Its surviving documentary context allows historians to link agreed cover to an actual adverse event. The evidence is particularly valuable because it makes the claims stage visible.

A second example, dated May 19, 1387, protected a shipment of 300 sacks of woad from Genoa to Valencia for 900 gold florins at a 3% premium. The 1387 Datini insurance description lists four underwriting interests in quotas of 300, 300, 200, and 100 florins. Taken together, the examples show differentiated premiums, division of cover among underwriters, identifiable routes, and the importance of commercial information. Neither these individual premiums nor the few named cases should be treated as representative average prices across all Mediterranean voyages.

Older literature sometimes cites 1347 as the date of the earliest authenticated marine-insurance contract. De Roover's 1945 documentary reassessment already described the uncertainty surrounding early forms. Later archival investigation and the 1343 evidence make an unqualified assertion that insurance first appeared in 1347 historically indefensible. A precise account must name the document, its archive or published edition, and the sense in which a date is claimed to be early. A record can be the earliest known example in one archive, the earliest identified specimen of a defined form, or the earliest transaction entered in a particular published dataset. These are not interchangeable statements.

The dates also demonstrate why chronology and functional analysis must be combined. A 1343 document written as a fictitious loan can contain the substantive promise of insurance even if its premium is absent from the written deed. A 1347 Palermo agreement with a clear premium is a stronger witness to overt pricing, but its explicitness alone does not mean that insurance began four years later. The surviving chronology supports a gradual, locally differentiated process rather than a single universally decisive birth certificate.

The Emergence of Distinct Premiums

A separate premium marks a major change in how the price of maritime risk appears in the historical record. In a sea loan, the merchant's expected cost of receiving capital is bundled with the creditor's exposure to a lost voyage. The promised repayment exceeds the advance, but the difference includes both the opportunity cost of capital and compensation for possible nonrepayment. It cannot be interpreted directly as a market price for insurance. In a separately contracted guarantee, the premium is consideration paid for protection over an identified exposure, while financing may be supplied by the merchant or another lender. This separation makes the cost of risk protection more observable, comparable, and independently negotiable.

The economic distinction can be expressed without anachronistic claims about medieval actuarial sophistication. Let L be the sum advanced by a sea lender and A the agreed repayment on safe completion. Where specified casualty losses extinguish the debt, the borrower receives an initial advance L and owes A only in the success state. The difference A minus L is not a premium for standalone cover: it is a combined contractual return on contingent credit. For a separate insurance agreement, a merchant pays premium P on an insured value V and holds a claim I(loss), subject to the agreement's terms. Financing remains outside this relationship. These symbols simply restate the contractual cash flows; they are not estimates from archives or a newly proposed financial theory.

The 1347 Palermo deed's 7.5% premium and the late-fourteenth-century Datini examples show that a recognizable insurance price was recorded in connection with value and voyage. Botticini, Buri, and Marinacci's transcribed medieval policy examples provide several concrete comparisons, including 6% on the 1384 Pisa–Barcelona venture and 3% on the 1387 Genoa–Valencia venture. These charges refer to particular coverage periods and journeys, not recurring annual premiums. Their different amounts should be understood in light of the vessel, shipmaster, cargo, direction, route, season, insured interest, and information available to underwriters.

The same archival study contains a sample of contracts for which explicit premia survive, permitting cautious statistical description. The reported average premium rates differ among the commercial centers: the study's premium-bearing observations report around 5.67% in Barcelona, 6.51% in Florence, 5.28% in Genoa, 7.70% in Palermo, and 2.46% in Venice. These are conditional sample averages from specific surviving documents, with very different observation counts, not comparable population estimates of each city's true long-run underlying maritime risk. The study's premium evidence and tables also reflect unequal observation periods, route mixtures, and recording conventions. The Palermo and Venice values in particular should not be turned into a ranking of medieval Mediterranean safety.

An explicit premium also creates the possibility of distinguishing the economic functions of several participants. A merchant supplies cargo and faces a concentrated exposure; an underwriter provides protection and acquires a claim to the premium; a broker may gather underwriting commitments and information; the notary or commercial scribe records the bargain; adjudicators enforce contested obligations. Differentiation makes specialized risk bearing visible in the documents. It can lower some transaction costs by allowing merchants to buy protection without changing their borrowing arrangements, while making price differences and disputes about covered events more transparent.

The relation between documentary explicitness and actual payment is fundamental. A fictitious loan may omit a premium even where protection was purchased. Such an omission limits what the record can establish about price, but it is not sufficient to conclude that the risk bearer gave away protection without compensation. Equally, the presence of a nominal sale price cannot be equated with an insurance premium when the transaction has not been shown to function as indemnification. Reliable comparison depends on what the recorded amount represents and when the parties exchanged it.

This is a point of genuine historical significance. A separately priced promise makes risk a contractual object capable of circulation beyond the lender–borrower relationship. It enables one merchant to finance goods, own the cargo, and procure cover through distinct counterparties. The result is an institutional expansion of possible commercial arrangements, whose actual reach still depends on counterparties' wealth, reputation, liquidity, and enforcement.

Risk Allocation, Pricing, and Underwriting Networks

An insurance contract reorganizes the financial consequences of an uncertain event. It cannot abolish the shipwreck, storm, seizure, or deterioration that gives rise to a claim. Medieval marine insurance likewise created an additional promise subject to the solvency and conduct of the underwriter. This distinction matters when interpreting the meaning of risk transfer. Transferring an exposure to another merchant reduces the insured's direct loss only insofar as the guarantor remains able and obliged to pay.

The earliest surviving agreements already display choices about the perimeter of protection. The Palermo 1347 policy included losses from maritime causes and human enemies, the itinerary from Palermo to Pisa, provisions for the cargo's loading and unloading, and a condition concerning justified deviations. Such details are reported in the archival reconstruction of the Palermo deed. Their presence means that the parties did more than refer generally to the uncertainty of navigation. They specified a recognizable object, exposure period, and pathway from an adverse event to a claim.

Human and natural hazards could be treated differently. Storms and navigational perils depended on seasons and waterways; piracy, warfare, raids, embargoes, and deliberate interference could change rapidly with political conditions. Late medieval trade networks carried information about ship departures, ports, captains, conflict, and commercial alliances. The medieval underwriting study finds evidence that human-generated peril influenced pricing. Its econometric analysis strengthens the assertion that insurers responded to observed commercial risks, although the underlying observational data do not demonstrate a universal or perfectly efficient premium formula.

The division of the 1384 and 1387 Datini placements among several named underwriters demonstrates another mechanism. The protected value could be allocated among distinct personal obligations. A merchant might thus avoid concentrating all counterparty exposure on one guarantor, and underwriters could participate in voyages without underwriting an entire cargo. Balard's analysis of Genoese coverage commitments similarly reveals the commercial and social networks behind apparently bilateral deeds.

Risk spreading was constrained by correlation. The loss of one vessel could simultaneously affect numerous cargoes and insured merchants, and conflict in a trading area could threaten many voyages together. A group of underwriters all tied to the same commercial center or to overlapping voyages would therefore face common shocks. The historical sources support the existence of divided obligations, not a calculation showing that these merchants had eliminated portfolio risk. That distinction remains central to interpreting claims about medieval diversification.

A broker's value emerged from maintaining knowledge of counterparties, assembling several commitments, negotiating rates, and navigating local contract conventions. These functions are economically intelligible in a market where information is valuable and contracts are imperfectly standardized. The historical evidence can establish brokerage activity and divided placements; it is far harder to prove the exact brokerage margin or quantify how much each intermediary reduced risk. The useful inference is institutional: independent insurance required channels through which a merchant could find suitable risk bearers, form an enforceable bargain, and pursue performance.

Commercial information was also unevenly distributed. A merchant might know more about cargo condition, the competence of a shipmaster, or the likelihood of departure; an underwriter might have better information about piracy or the willingness of a distant debtor to pay. Contract clauses specifying routes, timing, descriptions of goods, exclusions, and notice requirements can be interpreted as attempts to contain these practical information problems. Such interpretation should remain attached to particular clauses rather than treating every medieval merchant as a participant in a modern information-theoretic model.

Pricing records present related methodological difficulties. A stated premium rate is an observable contracted price, whereas the probability of a covered loss is usually not directly recoverable for the same journey. The difference between price and expected loss can encompass capital constraints, brokerage fees, private information, reputational concerns, enforcement costs, and negotiated market power. These components cannot be separately measured from one policy. Consequently, simple comparisons of the premium with a modern loss probability or with an annual interest rate produce misleading economic conclusions.

Underwriting networks also developed amid asymmetries of legal authority. A promise made in one town might cover goods shipped from another and litigated in a third. Merchant correspondence, notarial instruments, local courts, and shared conventions enabled such transactions, but enforcement was contingent on jurisdiction and the parties' standing. This is why the history of premium insurance is simultaneously the history of trade institutions. The promise could be purchased only when merchants had sufficient confidence that its meaning and performance would survive separation across distance and time.

Comparative Contract Analysis

The central comparison can be stated through three ordinary historical transaction types.

Dimension Contingent sea loan Fictitious insurance loan Explicit marine insurance
Initial capital advance Genuine lender-to-merchant advance Nominal advance described but ordinarily not made Insurance protection can be purchased without a loan
Main contingent obligation Borrower repays financier after stipulated success Risk bearer owes merchant upon stipulated loss Underwriter owes defined indemnity upon covered loss
Price in the source Contingent debt return bundles capital and hazard Premium may be concealed or recorded separately Premium is identified as the price of protection
Subject matter Advanced funds, collateral, and voyage Nominal debt referencing a vessel or cargo contingency Specified insured value or interest, perils, and voyage
Primary evidence problem Distinguish accident from borrower default Test factual absence of a genuine loan Test coverage language, rate, claims and underwriting shares
Typical archival witness Greek speech or Roman juristic text Genoese notarial deed Palermo policy or Datini private agreement

The table abstracts from particular source families and is intended for comparison; individual deeds may differ. It does not identify a uniform contractual path followed by every city. Its most important message concerns directionality. The contingent sea loan removes or limits a payment owed by the merchant after a maritime casualty. Fictitious-loan insurance and explicit insurance instead generate a payment or compensatory duty by a risk bearer after the casualty. The legal pathway can differ even where the insured merchant experiences a similar improvement in the loss state.

A simple hypothetical clarifies the financial relationship. A merchant purchases cargo for 1,000 units with personal capital. Without insurance, a total covered loss reduces the merchant's cargo wealth by 1,000. With a separate policy covering that full value at a 6% premium, the merchant pays 60 at the outset and receives a contractual right to 1,000 on a qualifying total loss, assuming full and timely performance. The net financial result depends on whether the cover is valid and the insurer pays; salvage, deductibles, partial loss, and other terms would modify the example. This calculation simply illustrates how a premium and coverage value interact; it makes no historical claim that every contract paid exactly this amount.

If instead a lender advances 1,000 at a 30% voyage-contingent return, the merchant receives the 1,000 initially and owes 1,300 after the specified successful voyage. If the covered loss extinguishes the debt, the borrower avoids repayment; the lender loses the advance. The lender has both financed and protected the merchant through one contingent-credit arrangement. It would be a category error to characterize the difference of 300 as an independently purchased insurance premium. The comparison is compatible with the observed voyage returns in Against Lacritus, though the numerical examples here are stylized.

The contractual differences generate different incentives. A contingent creditor may have a direct stake in vessel and cargo safety, because performance protects repayment. A separate underwriter can be concerned with representations about the vessel, shipment, route, and claims. Both may seek contractual safeguards against avoidable losses and fraud, but their practical monitoring opportunities and remedies may differ. The insured merchant, moreover, may purchase protection from several underwriters even if financing is concentrated with one lender or supplied from retained profits.

The historical comparison is strengthened by considering partial loss, loss caused by enemies, salvage, and delay. A contract might excuse repayment or require indemnity only for identified perils, while still enforcing duties arising from deviation, negligence, concealment, or nonpayment after safe arrival. A surviving document that contains only a short reference to risk may leave these questions unresolved. A more detailed policy improves the historian's ability to reconstruct the bargain but does not necessarily prove that its merchants had more sophisticated subjective risk beliefs than those who used another documentary form.

The development of separate insurance premiums can also be examined in relation to general average, a distinct maritime allocation mechanism under which participants in a voyage may contribute to an extraordinary sacrifice made for common safety. General average distributes a loss within a maritime venture according to commercial and legal rules. Insurance adds an external contingent counterparty who undertakes to protect a defined interest against specified perils or losses, potentially including a general-average contribution where the policy provides. These arrangements can coexist and should not be collapsed into a single early insurance history. The comparative maritime-risk history discusses the continued coexistence of general average, sea loans, insurance, and commercial finance.

In the same way, a partnership that shares profits and losses among investors is an arrangement for combining capital and business exposure, while a stand-alone policy sells a contingent financial guarantee to an insured interest. Medieval merchants used several of these mechanisms simultaneously. Their coexistence is an explanation of functional differentiation: a market gains flexibility when capital raising, enterprise ownership, transport losses, and external protection can be organized through different bargains.

Source Criticism and Unresolved Historiography

The strongest historical conclusions require care with survival, provenance, transcription, and classification. A notarial record proves that an instrument was drafted and preserved; it does not automatically establish that the promised consideration changed hands, that every voyage took place, or that all eventual claims were paid. An account-book entry can document a premium, but its purpose and accounting conventions may differ from those of the notarial deed. Litigation records demonstrate contested performance and provide detailed factual narratives, though adversarial statements may be selective. The proper evidentiary method is triangulation across genres rather than privileging one type of source without qualification.

The 1343 dating problem is a clear example. A Genoese Casanova document, a Palermo record, and later examples from 1347 are embedded in different source collections and scholarly traditions. Iodice's Genoese documentary history and Botticini, Buri, and Marinacci's archival data and contract readings support the conclusion that insurance-like agreements appear by the early 1340s in these interconnected commercial communities. They do not identify a verifiable first invented policy in world history. Claims of priority must therefore state whether they refer to the oldest presently surviving specimen, the earliest record in one market, the first document interpreted as a distinct insurance contract, or the earliest explicitly recorded premium.

A related issue concerns the word fictitious. It describes the economic departure from the formal loan or sale wording, not evidence that the entire contract was a forgery. A genuine instrument signed by real parties can use a legal fiction to express a real, enforceable obligation. The historian must determine what the fiction accomplished and which part of the transaction was real. Describing a 1343 deed as a fictitious loan does not make its historical existence doubtful; it signals the analytical gap between formal terminology and economic performance.

The Datini corpus illustrates both the opportunity and limits of large-scale records. The Italian state archival inventory of the Datini archive describes approximately 400 surviving insurance policies, including 402 documents associated with Pisa and Florence and documents dated to 1379 and 1385. It also provides the institutional context of the Datini enterprise's correspondence, account books, and geographic offices. The inventory's organization and the survival of a merchant network's records are not random samples of the whole Mediterranean economy; they reflect archival formation, the merchant's business interests, and later preservation.

Balard's Genoese sample and its chronology similarly raises a selection question. Many of the 1,169 documented acts belong to the later fourteenth century. This could represent increased transaction frequency, better survival, a change in notarial practice, or several factors together. A rising count of extant deeds cannot be assumed to measure an identical rise in the total prevalence of insurance. The observation that more insurance was recorded is more secure than a precise market-share estimate.

The distinction between printed editions and original manuscripts must be retained. The present research consulted a published translated legal code, digital editions of Greek and Roman texts, an accessible canon-law record, an official archival inventory, and modern scholarly historical analyses. It has not inspected original Genoese or Palermo notarial folios, conducted a fresh palaeographic transcription, or independently validated every named archival shelfmark. The detailed examples are therefore attributable to the historians who reported and analyzed them. This limitation bounds claims of originality but does not prevent a substantive independent interpretation of how the documented transactions differed.

Some historiographic assertions have stronger evidentiary support than others. The existence of conditional maritime loans in Greek and Roman records is securely documented. The presence of marine insurance in Italian commercial records during the fourteenth century is securely documented. The functional distinction between a true sea loan and an insurer's separate promise follows from the described payment conditions. The claim that canon-law condemnation directly caused the invention of insurance is more speculative: regulatory changes can influence contract forms without constituting the sole or demonstrable origin of a business practice. The claim that insurance began at precisely one identifiable date is still weaker.

An additional analytical caution concerns colonial and non-European histories. The Mediterranean evidence in this monograph is informative about the emergence of identifiable premium marine-insurance agreements in a documented European setting. It supplies no warrant for assuming that other regions lacked elaborate risk-sharing mechanisms before the fourteenth century. The terminology of world origins is especially vulnerable to exaggeration when the archival record is geographically imbalanced. A historically responsible account states the limits of the particular document category and region being studied.

Legal transmission creates its own risks. Gregory IX's Naviganti in the Liber Extra must be identified by its canonical position rather than relying exclusively on secondary shorthand about the year 1236. Later authors can reproduce chronologies whose legal referents differ. A source-critical interpretation should say whether it describes the enactment of a rule, compilation in a canonical collection, survival of a manuscript, or later scholarly dating. Similar precision applies to Roman juristic texts transmitted through Justinian's sixth-century compilation: the date of a jurist's underlying opinion differs from the date of the Digest's compilation.

The historical record nonetheless permits a robust account of institutional change. The documents independently attest contingent repayment, legal restrictions on profit from loans, notarial fictions, separate insurance payments, multiple underwriters, precise insured interests, and claims arising from losses. Their interaction supports the conclusion that insurance emerged as a differentiated contractual service within medieval commercial markets. It does not require the invention of a new untested theory or a claim that every observed practice can be reduced to a single causal law.

Historical Conclusions and Contemporary Relevance

The transition from sea loan to separately priced policy shows how a practical financial function can pass among different legal forms. Ancient and classical maritime loans combine finance with the lender's assumption of voyage risk. Medieval canon law affects the legitimacy of returns from this kind of credit. Genoese notarial strategies use ostensibly familiar transactions to record protection whose performance is increasingly directed toward compensation for loss. Palermo and Tuscan documents make the premium, the insured exposure, and the identities of risk bearers more directly observable. Across these settings, the changing location of the contingent obligation is the most reliable indicator of differentiation.

This sequence is historical rather than a universal linear formula. Maritime loans continued to coexist with insurance, partnerships, guarantees, and general-average allocations. Contractual variants differed by region, commercial network, and forum. The Genoese experience examined by Zanini (2023) demonstrates that loan financing and other techniques persisted in later maritime business. The co-presence of instruments makes sense because merchants required several services: capital, shipment, protection, enforceability, and settlement. Insurance supplied one of these without rendering the others obsolete.

The transformation is also important for the history of the price system. When a marine risk payment becomes separately observable, merchants can compare terms for protection independently of the cost of borrowing. More than one risk bearer can participate in a placement. Brokers and information networks can develop around the specific service of allocating uncertain loss. Legal disputes can increasingly concern whether a covered peril occurred and what compensation is due, rather than whether a purported loan should be repaid. These changes are documented in the surviving medieval records to different degrees; the broader market mechanism is an analytical synthesis grounded in them.

The value of distinguishing finance from insurance remains evident in modern legal economics. A creditor may finance a cargo while the borrower buys separate insurance on its maritime interest, and the creditor may also require loss-payee or collateral arrangements. The separation allows credit risk, insured physical risk, and counterparty performance risk to be identified and managed through different contracts. This modern analogy clarifies the historical transformation; it does not imply that a medieval merchant operated within contemporary insurance regulation or accounting systems.

There is an equally important lesson about documentary labels. Contracts must be read as structures of rights and duties, not simply as the nouns by which parties or historians name them. The absence of a visible premium does not establish the absence of a risk price where the instrument deliberately masks consideration. Conversely, the presence of an unusual debt repayment condition is insufficient to show that a third party has sold independent insurance. Examining initial cash movements, loss contingencies, remedies, pricing, and document context provides a more dependable comparison.

Finally, the evidence supports an interpretation of institutional innovation as a cumulative commercial achievement. Legal and religious constraints contributed to changing forms; access to merchant capital made unbundled protection useful; notaries and brokers made transactions more reproducible; courts supplied remedies; information networks supported pricing and counterparty choice. No single factor is demonstrated to have been both necessary and sufficient. Together, the sources show how a distinct market for promises about losses became visible and institutionally durable.

Conclusion

Sea loans and marine insurance belong to a connected history of allocating the uncertain financial consequences of maritime exchange. Their difference can be traced through actual contractual performance. A contingent sea loan finances a venture and places a defined part of the voyage risk on the creditor. A separate insurance bargain transfers a specified loss exposure in exchange for a premium without necessarily supplying venture capital. A medieval instrument can carry the economic substance of insurance under a fictitious loan or conditional sale, so the chronology must be reconstructed from rights, cash flows, conditional duties, and evidence of payment rather than from titles alone.

Greek maritime credit records and Roman juristic writings establish that merchants and legal systems recognized contingent sea finance well before the surviving medieval insurance market. Canonical restrictions on usury influenced the vocabulary and legal acceptability of commercial returns. Fourteenth-century Genoese documents reveal the inversion of familiar loan and sale forms to express a loss-contingent promise. The Palermo and Datini records provide concrete evidence of separate premia, differentiated underwriting shares, identifiable voyages and perils, and claims procedures. Their preservation reflects particular archives and therefore does not reveal a unique invention date for the world.

The historical significance of the transition lies in the separation of risk protection from the advance of capital. This enlarged the range of contractual arrangements available to merchants and enabled marine risk to be priced and borne by counterparties specializing in that obligation. The documents support a multi-causal account involving religious and civil legal norms, merchant capital, documentation, information, adjudication, and the scale of long-distance trade. The analysis is an original comparison of established sources and does not claim a novel standalone theory or newly discovered primary manuscripts.

The central research finding is thus a contractual and institutional one: the origins of recognizable premium marine insurance lie in the development of a separately enforceable, independently priced promise to respond to a defined maritime loss. Ancient and medieval sea loans provided crucial antecedents and parallel mechanisms, while the surviving evidence of the fourteenth-century Mediterranean demonstrates the emergence of that distinct promise into an increasingly organized commercial practice.

References

  • Balard, Michel. (1978). “Assurances et commerce maritimes à Gênes, dans la seconde moitié du XIVe siècle.” Annales de Bretagne et des Pays de l’Ouest 85(2): 273–282. Full text and DOI.

  • Botticini, Maristella, Pietro Buri, and Massimo Marinacci. (2023). “Presidential Address 2023: The Beauty of Uncertainty: The Rise of Insurance Contracts and Markets in Medieval Europe.” Journal of the European Economic Association 21(6): 2287–2326. Publisher and DOI.

  • Cecchi Aste, Elena, ed. (2004). L’Archivio di Francesco di Marco Datini: Fondaco di Avignone. Inventario. Pubblicazioni degli Archivi di Stato, Strumenti CLXIII. Rome: Ministero per i Beni e le Attività Culturali, Direzione Generale per gli Archivi. ISBN 88-7125-250-0. Official archival inventory.

  • De Roover, Florence Edler. (1945). “Early Examples of Marine Insurance.” Journal of Economic History 5(2): 172–200. Publisher and DOI.

  • Demosthenes, attributed. Against Dionysodorus (Oration 56). Ancient Greek forensic text; modern translation in A. T. Murray, Demosthenes, Private Orations, Loeb Classical Library. Perseus catalog and text entry.

  • Demosthenes, attributed. Against Lacritus (Oration 35). Ancient Greek forensic text; A. T. Murray's English translation presented by ToposText. Text and translation.

  • Gregory IX / Decretales Gregorii IX (Liber Extra). (1234, compilation). Book 5, title 19, chapter 19, “Naviganti”; printed canonical edition in Emil Friedberg, Corpus Iuris Canonici, vol. 2 (Leipzig, 1881). Digital source record.

  • Hammurabi. Code of Hammurabi, especially provisions 102–105 (Old Babylonian period). English translation by L. W. King, reproduced in the Yale Law School Avalon Project. Translated text.

  • Iodice, Antonio. (2023). “Marine Insurance in Early Modern Genoa (1564–1571): A Risk-Shifting or Risk-Sharing Tool?” Asia-Pacific Journal of Risk and Insurance 17(2): 215–231. Publisher and DOI.

  • Justinian I, comp. (533). Digest (Digesta seu Pandectae), title 22.2, De nautico fenore; legal excerpts from Modestinus and Papinian; edited by Theodor Mommsen and Paul Krueger, Latin online corpus, Université Grenoble Alpes. Latin text, Digest 22.2.

  • Fusaro, Maria, Andrea Addobbati, and Luisa Piccinno, eds. (2023). General Average and Risk Management in Medieval and Early Modern Maritime Business. Cham: Palgrave Macmillan / Springer Nature. Publisher book record.

  • Zanini, Andrea. (2023). “Financing and Risk in Genoese Maritime Trade During the Eighteenth Century: Strategies and Practices.” In General Average and Risk Management in Medieval and Early Modern Maritime Business, edited by Maria Fusaro, Andrea Addobbati, and Luisa Piccinno, 335–359. Cham: Palgrave Macmillan. Open-access chapter and DOI.