5.1.1 Habitat Destruction

One long-familiar medieval economic trend needs mere re-articulation in ecological terms: growing reliance on cereal food meant permanent plowed fields replaced woodland from central Spain to Sweden and Wales to Poland.Footnote ³ Clearance of forests, which slow runoff and maintain steady streamflow, inescapably alters the pattern of stream discharge to greater seasonality and irregularity. Rain and meltwaters run more quickly off farmland. Larger and faster runoff more forcibly abrades stream beds and channels, and then falling water levels leave a contracted stream and deposits of eroded materials. One astute observer in late thirteenth-century Alsace noted how clearance of the Vosges in his own lifetime had caused much more rapid and dangerous runoff.Footnote ⁴ Modern scholars detect like sequences of medieval deforestation and flooding in the Po basin and in central Poland.Footnote ⁵ Biologists now also know that unstable flow regimes make life hard for fishes. Those living in running water must expend more energy during floods. They lose eggs and young to winter spates and suffer high mortalities when small streams dry up in summer. Species which spawn in flooded margins are adapted to consistent seasonal patterns of rising and falling water, so instability disrupts their reproductive behaviour.Footnote ⁶

Soil erosion and alluvial deposition is becoming a well-known consequence of medieval agricultural expansion. First the natural vegetative cover was removed. Then characteristic medieval farming practices disrupted the soil surface and its structure. Plowmen drove long straight furrows or pulverized the soil to prepare for autumn sowing of winter grains. Bare fallow stripped all plants from a third or a half of arable almost year-round. Large open fields broke neither wind nor water nor the creep down slopes of the soil itself. All this let topsoil flow to the watercourses, especially during heavy winter rains and snowmelt.Footnote ⁷ In the Leine valley of Saxony, where during the 780s and 790s Frankish conqueror Charlemagne had promoted new rural settlements and clearances, large sediments from eroded topsoil overlain with former subsoil yield radiocarbon dates between the 790s and 850s. Across the entire lower Rhine valley a major pulse of sedimentation set in around 1000 CE.Footnote ⁸ In southern England, where woodland clearance got under way in the eighth century and was complete by the eleventh, erosion and deposition rates in the river Nene and the upper Thames accelerated from around 800 to reach in late Saxon and early Norman times all-time maxima tenfold their prior Holocene average.Footnote ⁹ Among the hills of Dauphiné about the year 1000 farmers pioneered the wooded shores of Lac Paladru; their new grain fields quickly produced higher rates of erosion now visible in well-dated lake-bottom sediments suddenly full of loam from plowed topsoil and organic waste from cattle.Footnote ¹⁰ At the crest of medieval expansion in the early fourteenth century, researchers both along the French Alps and across central Europe find unusually heavy erosion episodes when climatic change brought heavier precipitation to precisely those landscapes most recently deforested and intensively plowed for maximum output of cereal grains.Footnote ¹¹ Incrementally, region by region, watershed by watershed, surges of hydrographic instability, soil erosion, and deposition followed pulses of local conversion of woodlands to arable.

Soils from upriver washed down to the estuaries. The Oude Rijn mouth of the Rhine in Holland silted shut by the eleventh century, and between the thirteenth and sixteenth centuries the expanding delta of the Wisła filled in the one-time bay between Gdańsk and Elbląg.Footnote ¹² Long unavailing struggles of Bruges against plugging of the Zwin, and of Ravenna against filling of the Po delta are historical commonplaces.Footnote ¹³ For some fish species, this could mean loss of important migratory, spawning, and juvenile habitats. Heavy silt loads make water more often turbid, reduce light penetration, and can smother fish or prey species adapted to live in weed or gravel beds. Like the more dramatic alternation of floods and low water, the effects favour certain species relative to others. A tenth-century monastic chronicler at Novaliense in the Italian Piedmont was well aware that clear mountain water held lots of fish and a muddy stream few.Footnote ¹⁴ Soils from unstable cleared lands in tenth–twelfth-century Sicily went down local rivers to trigger shifts in their fish populations.Footnote ¹⁵

To process the new grain supplies a little-used late antique invention, the watermill, surged across the medieval landscape. From perhaps a couple hundred in King Alfred’s England, they multiplied to 5,624 in the Domesday Book of 1085. In Poitou, Berry, Languedoc, Burgundy, and Lorraine mills proliferated from the tenth century through the twelfth. On the Aube, where fourteen mills are recorded in the eleventh century, sixty-two may be counted in the twelfth century, and almost two hundred in the early thirteenth. Then and later their construction became a normal part of rural development in the Egerland and in central Silesia. Late thirteenth-century Milanese writer Bonvesin de la Riva estimated the territory of his city held 900 mills running about 3,000 wheels.Footnote ¹⁶ One historian of technology summed it up this way:

In fact, as remarked in Chapter 1, stream ecosystems had long existed in most of Europe, but less often ponds. Medieval watermills commonly drove their overshot or breast wheels by using a dam or weir two to five meters high to concentrate the falling water and pond a reserve supply of it. Medieval millwrights learned to do this on ever larger rivers. Once a design was in place at a location, it rarely changed.Footnote ¹⁸ Dams blocked running water and created still water: each of the nineteen mills erected during the Middle Ages on the forty-kilometer-long Skrwa, a Masovian tributary of the Wisła, had a dam about three meters high and a pond covering up to ten hectares.Footnote ¹⁹ As moving water slows, it drops the solids it has carried in suspension. On the Derwent in the English Midlands two meters of gravel and silt alluvium eventually covered a one-meter timber mill dam, gate, and race dated by dendrochronology to the mid-twelfth century.Footnote ²⁰ The broad surfaces of standing waters absorb more solar energy. This both warms the water and further improves conditions for growth of rooted plants. In twelfth-century Picard charters and conveyances slower, deeper, and weedier waters backed up behind mill dams and weirs all along the Scarpe, the Oise, and the Somme.Footnote ²¹ Ubiquitous watermills formed and multiplied a new kind of aquatic habitat, one to be probed more deeply below.

On existing watercourses and their native fish populations mills had immediate effect, for they blocked movement of migrants. Like the concentrations at natural barriers, those at dams and weirs offered fishers profitable access to migratory species. The contemporary biographer of St. John of Metz (Gorze, d.974) actually thought his monks’ need for fish was why he built mill dams. Even deep in central Saxony, operation of a salmon trap at the mill dam at Lauenheim on the River Zschopau fueled a century of dispute (1293–1393) between Altzelle abbey and the von Steinbach family.Footnote ²² Possession of mills was associated with the right to take eel on the Duero in Castile, the Garonne near Toulouse, the Meuse around Liege, and in the early fourteenth-century psalter illuminated for Sir Geoffrey Luttrell of Lincolnshire (revisit Figure 3.7).Footnote ²³

But impassable barrier dams kept migratory species from vital spawning habitat. Blocked runs of fish – were they trout or shad? – ascending the Sarca from Lake Garda in 1210 caused the bishop of Trento, who held sovereign fishing rights in that county, to require removal of mill dams at Arco.Footnote ²⁴ For the sake of the salmon Scottish king William (1165–1214) established judicial precedents requiring all dams and weirs be fitted with a permanent mid-stream opening and all barrier nets be lifted from each Saturday evening until Monday sunrise.Footnote ²⁵ An English law book from the 1290s, Fleta, likewise acknowledged that mill dams could damage established fisheries.Footnote ²⁶ Dutch historical ecologists have most recently argued from sparse medieval and early modern salmon price series as well as material evidence at prehistoric and medieval sites for a correlation between construction of water mills in the Rhine basin and the first declines of salmon populations there. Others have asserted that eleventh–twelfth-century dikes built to drain marshes suppressed sturgeon stocks in the Rhine delta.Footnote ²⁷ Impassable dams are well known to break the ecological continuity of rivers and fragment even populations of resident fishes, while other modern research indicates a succession of even modest two- to ten-meter barriers has cumulative negative impact on upstream populations of migrants.Footnote ²⁸ These losses mattered because, ecologists agree, the spawning environment in fresh water determines the productivity and survival of anadromous fish.Footnote ²⁹

Turning from rural development to other aspects of the medieval economy, human population growth and urbanization in the typically organic-based preindustrial resource system affected both water chemistry and hydrological conditions for aquatic life. Waste from more and larger human concentrations, from rural monasteries to towns of twenty or even fifty thousand, necessarily increased the nutrient load – i.e. soluble nitrates – in watercourses.Footnote ³⁰ The several hundred monks and lay brethren at early thirteenth-century Clairvaux were served by a diversion of the Aube river, which ran through gardens, mills, brewery, fulling mill, tannery, laundry, and latrines before rejoining the main stream.Footnote ³¹ Such point source pollution also typically flowed from elite lay residences.Footnote ³² Local streams likewise received the human, animal, and craft waste of towns, whether by runoff from street disposal (even with intentional diversions to flush gutters as at Milan, Strasbourg, and Goslar), by purposely emptying the contents of cesspits into flowing water below town (Köln), or by direct siting of latrines over watercourses (Rouen, Nürnberg). If, as inhabitants of the Terra Firma legitimately jibed, “I Venexian caga in aqua,” there was yet another source of nutrients for the lagoon.Footnote ³³ No wonder contemporary Italian doctors and dietitians doubted the wisdom of eating fish from waters polluted with urban effluent.Footnote ³⁴ But Robert Guillerme has argued that organic acid- and alkaline-based processes used by early medieval textile and leather crafts caused “the precipitation of solid organic materials in water which river currents carried beyond city limits.”Footnote ³⁵

And what was downstream? One introductory fish tale has already told of Constance’s pollution of its lakeshore with urban wastes. Along what was London’s little river Fleet, sediments from a human generation or two of the mid fourteenth century show loss of molluscs requiring clean water and appearance of diatoms typical in dirty water.Footnote ³⁶ Excavations in the bed of the Pegnitz below medieval Nürnberg recovered late medieval butchery waste and household refuse, findings which corroborate the stream’s foul repute when each summer’s low water left it long unflushed. By the early 1400s Parisian effluent was likewise making the Seine below town “infectée et corrumpue” every summer.Footnote ³⁷ All are symptoms of aquatic ecosystems under stress.

Nor, despite Guillerme’s optimism, did medieval industry merely add nutrients. More immediate toxic effects came from crafts such as slaughtering livestock, tanning, or extracting fibers from flax and hemp by wet decomposition (‘retting’). When the latter activity killed fish near Douai in 1452, holders of fishing rights sued a clothier for damages.Footnote ³⁸ Brewers, fishers, and ordinary consumers at Colchester in 1425 complained that the tanners and tawers caused the “impayring and corrupcion” of the river Coln and “destruction of the ffysche therynne.”Footnote ³⁹

Toxic heavy metals emitted from medieval mining and metallurgical processes still contaminate substrates, riverbanks, and floodplains in widespread European watersheds. Even now lead concentrations in waste deposits from medieval mines in the Pennine headwaters of the Tyne, Don, and Ouse as far downstream as York exceed those of Roman or modern industrial dates.Footnote ⁴⁰ Maximum contamination in lakes and shorelines of the Harz and the Staufer basin of the Schwarzwald from tenth- through thirteenth-century processing of copper, lead, zinc, and cadmium is as much as ten times the modern legal limit.Footnote ⁴¹ Alas, I know of no analyses for heavy metals in medieval human or fish remains from these and other affected regions, but studies of human skeletons elsewhere clearly indicate generally high lead exposure among especially European urban populations of high and late medieval date. Even the avid sixteenth-century mining promoter Georg Agricola had to admit to the deadly consequences of mining and refining.Footnote ⁴²

Activities in the urban and commercial sector further impeded the free flow of water. Since the eleventh century castles and towns had diverted rivers to fill defensive moats; this stratagem gained popularity in the later Middle Ages.Footnote ⁴³ Accessible markets for fuel drove extensive peat-digging which created the Norfolk Broads, smaller but more numerous plassen in North Flanders, and South Holland’s vast but equally anthropogenic Haarlemmermeer between Haarlem and Leiden.Footnote ⁴⁴ And between the eleventh and fourteenth centuries still more weirs, dams, and ponds, were built to power new industrial operations like malting, fulling, metal-working hammers and bellows, sawmills, and paper-making.Footnote ⁴⁵

The type and scale of physical and chemical changes which medieval economic development brought to European inland waters most directly and heavily affected small- and medium-sized watercourses. Brooks, streams, and small rivers are by their very size, high ratio of surface area to volume, and abundance in the landscape more closely tied than large rivers to their immediate terrestrial environments. Removal of bankside vegetation; local ditching, diversion, or embankment; small mill ponds and dams; and effluents from concentrations of livestock or humans have profound local impact, removing the whole waterway from its natural form and sources of energy. Large rivers, in contrast, are linked to their surroundings through their multiple channels and extensive floodplains, so simple and local changes in riparian conditions have less effect.Footnote ⁴⁶ Hence the impact of preindustrial economic development differed in degree and kind from that of industrial development. Yet the finding should not be oversimplified. The much-studied Rhine and its major tributaries, for instance, are said to have suffered little from human activity before channelizing and embanking in the early nineteenth century began a total degradation.Footnote ⁴⁷ But most studies of major European rivers proceed from a present and retrospective standpoint with little deeper historical knowledge or awareness. The evidence of medieval landscape change is overwhelming. Watersheds are systemic continua; what enters at the top flows all the way down. Besides dams cutting off what had once been the highest spawning sites, medieval deforestation and erosion must be acknowledged a principal cause of more erratic flow regimes and contributor of material to the many mainstream sandbars and islands known from early modern records.

5.1.2 Perceptions of Overfishing and the Evidence of Depletion

For many past and present observers fishing is the most obvious and understandable human impact on local and regional aquatic ecosystems. Medieval efforts to satisfy demand for fish as food rose with and beyond rising human numbers. Widened adherence to Christian food rules encouraged fish consumption. We have seen the growth and operation of urban fish markets all across medieval Europe. Such markets and the professional fishers who supplied them were by the thirteenth century held mainly responsible for rising exploitation of fisheries in northern Flanders.Footnote ⁴⁸

Populations of fishes which Europeans liked to eat came under evident stress. One sign is contemporary awareness of damaged resources, commonly conceived as overfishing. A remarkably explicit early articulation prefaces the first full-scale fisheries ordinance issued by French King Philip IV in 1289:

Philip’s remedy is examined in Chapter 6 below. A more particular case of fishing pressure is reported from the Pinzgauer Zellersee, high in the Salzburg Alps. Its rich fisheries drew a mid-fourteenth-century settlement of professionals, who paid the archbishop 27,000 whitefish and 18 lake trout a year for the right to take, smoke, and sell still more. After one human generation the whitefish catch collapsed, and replacement stockings of pike ate nearly all the trout, so the fishing community determined to rest the lake for three years and then to fish only with far fewer nets in a limited season and a restricted area.Footnote ⁵⁰

By the late fourteenth century petitioners blamed weir fishing for decline of salmon and sturgeon in the Thames estuary, and English coastal fishers a generation later conceded they had depleted local stocks – to excuse, it should be noted, their illegal fishing elsewhere.Footnote ⁵¹ About the same time Siena legislated against overfishing in the lagoon of Orbetello and authorities at Santander tried to deter local depletion by taxing nearshore catches.Footnote ⁵² Perceptions of damaged resources were by no means confined to inland waters.

Less subjective indicators allow closer description and diagnosis of localized fish stocks under stress. Our opening tale of the sturgeon revealed a steady reduction in size of those eaten at tenth- through thirteenth-century Gdańsk. Like trends are more vaguely visible in some other large central European fishbone assemblages predating 1200.Footnote ⁵³ At Abbeville on what was initially the Somme estuary the share of large adult flatfishes at excavated sites fell from the twelfth century to the early fourteenth, while that of small, mainly immature, specimens rose. Those reciprocal trends reversed after mid-century human population losses, but had resumed by 1500.Footnote ⁵⁴ As someone in the entourage of Philip IV already seemed to know, shrinking average size indicates a stock where more fish are being extracted before reaching their full growth.

Shifts in species composition of medieval catches are still more telling. Investigators have diagnosed a decline of anadromous and cold-water varieties during the high and later Middle Ages in several regions of western Europe. Here our sad introductory tale of the sturgeon needs only brief review. Zooarchaeologists studying Baltic sites concur in the steady decline not only in size, but in numbers and relative frequency of this taxon between the seventh/ninth centuries and the twelfth/thirteenth and point to overexploitation (Figure 5.1). Records around the North Sea trace a diminishing presence, too, although some observers argue more for loss of habitat. The European sturgeon may better have survived in the long-fished Mediterranean, although market price lists less and less often included it. From a common though costly food item, sturgeon everywhere became a rare symbol of prestigious luxury.

Figure 5.1 Sturgeon in fish remains from tenth–fourteenth-century Gdańsk.

Data as published in Makowiecki, “Exploitation,” fig. 7, and “Usefulness,” fig. 3, p. 109. Graph © R. Hoffmann.

Atlantic salmon, like sturgeon, long served to display a host’s high status as ubiquitous features of conspicuous consumption at festive elite banquets, even well away from the sea.Footnote ⁵⁵ As salmonid remains preserve poorly in archaeological contextsFootnote ⁵⁶ it is worth observing the presence of salmon bones in excavations from Anglo-Saxon Wraysbury in Berkshire, the local Slavic prince’s dwelling at high medieval Hitzacker on the Elbe, twelfth-century castles along the lower Rhine, the neighbourhood of the late medieval Louvre palace, and a contemporary house of canons at Saarbrucken.Footnote ⁵⁷

By the 1200s good anecdotal and other evidence across much of western Europe indicates decline of natural salmon stocks in especially smaller rivers and upper tributaries. Contemporaries attributed the depletion to barriers and competitive overfishing, but habitat changes resulting from agricultural clearances are also implicated. Historian Angelika Lampen traced the collapse of salmon in archives of the convent at Werden on the Ruhr from abundance in the eleventh century to absence in the fourteenth.Footnote ⁵⁸ Fears of damaged salmon runs were voiced in mid-thirteenth-century Northumbria and complaints of weirs and illegal fishing killing smolts and damaging runs could soon be heard on the Thames, Severn, Wye, and Meuse.Footnote ⁵⁹. On the small coastal rivers of lower Normandy, where plowed fields and watermills multiplied through the twelfth and thirteenth centuries, the generous annual gifts of salmon offered in eleventh- and twelfth-century charters from reliable local catches were distinctly fewer by 1300. Continued ‘overfishing’ thereafter brought near-total destruction of the runs by the mid-1400s, with local records referring rather to gifts of individual salmon imported from Ireland and Scotland.Footnote ⁶⁰ From the fifteenth-century start of suitable records along the middle Rhine the numbers and weights of salmon there taken go steadily downwards.Footnote ⁶¹ Many river basins saw by the later Middle Ages the pressure on salmon shift down to estuaries, making it out to be a marine rather than inland fish.Footnote ⁶² Even in wealthy Parisian households and prosperous Flemish monasteries consumption of once-favoured sturgeon, salmon, trout, and whitefish shrank to undetectable by around 1500.Footnote ⁶³ Yet despite general late medieval increase in consumption of marine fishes, salmon retained high cultural significance and commensurate prices across their entire northern and western European range.

Like the sturgeon, then, Salmo salar is a prime example of widespread localized medieval human impacts on European riverine, notably anadromous, fish populations. European fishers and consumers were well aware of these losses long before modern industrialization.

5.2.1 Eel

Eel are as migratory as salmon but travel in reverse directions (catadromy). Unlike the sensitive eggs and young of salmon, subadult eels enjoy broad ecological tolerances and omnivorous habits during their long maturation in estuaries and waters far inland. With well-chosen techniques people could catch eels throughout their freshwater phase and when the sexually mature adults migrated downstream to spawn at sea. Despite cultural antipathies to its snake-like morphology and benthic habits,Footnote ⁶⁵ this species evidently supported long-standing fisheries and human consumption in Atlantic and Mediterranean (but not Danubian or other Black Sea) drainages. Until sieving became standard archaeological practice, tiny bones and high oil content made eel remains evasive, but regular fishing of eel at weirs, mills, and in stillwater habitats everywhere supported heavy local consumption of fresh and lightly processed catches throughout the earlier Middle Ages.

Human activities described to start this chapter greatly enlarged nutrient-rich stillwater habitats, with the unplanned result of much favouring the eel and soon those who could fish for it. In precociously deforested England as early as the seventh- through mid-ninth centuries, eel was the most common taxon in all fish remains recovered from multiple contexts at York and in elite settlements at Canterbury and Lyminge, Kent. In decades around 1000 it remained as abundant at York, though yielding primacy to herring, and there continued in that second rank until the 1200s.Footnote ⁶⁶ Further inland in Aelfric of Eynsham’s time (c. 1000) eel were the single most common fish eaten in his monastery and from the twelfth century there, too, remained second only to preserved herring.Footnote ⁶⁷

Contemporary fish remains from northern France and the Low Countries point to parallel developments. Eel reached two-thirds of remains from twelfth-century Deventer, but thereafter fell back to only one-third behind undifferentiated small cyprinids.Footnote ⁶⁸ In the Somme basin eel emerge as the most common fish species from the ninth century with special abundance in urban contexts at Amiens through the twelfth and thirteenth. Although herring later steadily gained pre-eminence there, even as late as 1449 Cistercians at Gard abbey on the Somme were handling in single transactions as many as 13,000 eel. In food waste which fell through gaps in the kitchen floor in the contemporary Benedictine abbey at Ename outside Oudenaarde, eel remains at 12.6 percent of all fish (and 20 percent of those locally acquired) were also outnumbered only by herring.Footnote ⁶⁹ Much further south in kitchen waste from the priory at Charité-sur-Loire eel ranked second only to small native cyprinids from the eleventh century to the sixteenth, when those monks turned to carp as the main fish on their menus.Footnote ⁷⁰

Across the whole Rhine delta region of Holland and Flanders, a great and long-term expansion of shallow estuarine and freshwater lakes since at latest the 1200s resulted from subsidence of drained peat lands, peat mining, diversion of rivers, rising water levels, and both local- and large-scale wave action. These turbid, fertile, and accessible lentic habitats soon supported large and lucrative commercial fisheries for eel. Sluices of the Spaarendam beside Haarlem in the late 1400s annually yielded 150,000 eel (ten to twenty tonne). Fishers supplied towns in the region and even exported to England.Footnote ⁷¹ Where anthropogenic environmental change came later than it had further west, as in Poland, eel populations and human use of this fish may only more slowly have expanded inland.Footnote ⁷²

Western Mediterranean watersheds show similar trends. In the Rhône delta eel replaced sturgeon as the principal fishery by the fourteenth century, but already in the twelfth residents were alert to the lucrative returns possible from active fishing and storage of live eel in enclosed ponds or channels.Footnote ⁷³ Energetic Italian pursuit of this species got under way then, too. Local eel fisheries spread along Tyrrhenian shorelines from Tuscany to Sicily, into the Adriatic around the mouths of the Po, and in natural lakes and marshy river valleys of the interior, all locations where other kinds of evidence indicate increased lagoon formation, siltation, and eutrophication.Footnote ⁷⁴ So great was demand that from at latest 1275 each autumn the commune of Perugia sponsored transfers of juvenile eel (elvers) from the Chiana river into Lake Trasimeno and continued to do so even after Pope Martin IV died there in 1285 after eating [too many?] eel. During the 1350s and 60s the papal-owned fishery at the outlet of Lake Bolsena continued to ship thousands of eel a year to the curia then in Avignon.Footnote ⁷⁵ Eel matched sea breams as the most common fish taxa eaten at a late fourteenth-century palace in Tarquinia, just up the coast from Rome.Footnote ⁷⁶ We can but speculate on the point of balance for medieval eel stocks between anthropogenic enlarged habitats and intensified human predation.

5.2.2 Herring Fisheries on the Rise

People fishing for and eating herring enter the historical record of Atlantic (including North Sea and Baltic) coastal communities during the early Middle Ages.Footnote ⁷⁷ Then with simultaneous rising human numbers and environmental pressures of the tenth through thirteenth centuries these activities greatly increased in scale and distribution into nearby inland districts. Herring bones are as fragile and elusive in unsieved archaeological contexts as are those of eel. (Figure 5.2) Though pelagic in habits, schools of these plankton eaters once frequented close inshore waters off northwestern European beaches and penetrated deeply into now long-obliterated estuaries there.

Figure 5.2 Archaeological herring bones, unsorted. Herring and other bones from the Blue Bridge Lane site, York, fourteenth century.

Photograph © James Barrett. Used with permission.

From the south coast of the Baltic all the way around to the English Channel seasonal spawning concentrations offered local opportunities for rich catches within sight of land.Footnote ⁷⁸ Reproductively isolated estuarine stocks could be taken in fixed traps and weirs and those along open shorelines from small boats with light seines and drift nets, especially at night. Writing about 1200, Danish chronicler Saxo bragged that the fish along the Scanian coast were so abundant that they blocked shipping and could be caught by hand.Footnote ⁷⁹ But these oily creatures spoil in a day unless promptly dusted with salt, smoked, or packed in simple salt brine. Such light cures make herring – ‘powdered,’ ‘red,’ or ‘pickled’ – an inexpensive portable food, palatable enough for several months, especially during the cold season after the autumn spawning. By the mid-twelfth century coastal artisans from Picardy to Pomerania were tapping the silvery billions to feed themselves, their neighbours, and nearby inland populations, especially in fast-growing towns.

The booming twelfth-century herring industry had emerged from a historically obscure century and more of parallel development by fishers with access to local consumers and local fish. Early medieval coast-dwellers from Sussex to Sweden ate herring they, their neighbours, or their subjects caught from local shoreline and estuarine stocks.Footnote ⁸⁰ By about 1000 these catches were also supplying nearby inland consumers.Footnote ⁸¹ At Haithabu on the Schleswig isthmus, where Viking traders gathered from about 800 until 1066, this species contributed 38 percent of the 13,842 identified fish bones. The oldest medieval herring remains in inland Flanders appear only in the late tenth/early eleventh century and those only in the mercantile settlements at Ghent and Ename.Footnote ⁸² Until this time herring everywhere looked like a traditional subsistence or artisanal fishery being practiced seasonally along seashores close to consumers.

The eleventh century was then plainly transitional at many places around the herring coasts, with remains of that species at York, London, and elsewhere in eastern England a principal marker for archaeozoologist James Barrett’s ‘Fish Event Horizon,’ the significant appearance of marine fishes in local diets.Footnote ⁸³ Some of the first strong evidence of heavy commercial use comes from the Pomeranian coast of the Baltic and southern shores of the North Sea. Before 1100 inland Poles were well aware that the beaches between the Odra and Wisła estuaries were full of fresh herrings and a source for salted ones. The rich representation of herring in food remains from coastal sites thins out toward the interior. Still, from Pomerania likely came the herring bones found in some strongholds of the emergent Polish state along the Warta c. 1000 and more commonly in eleventh–twelfth-century layers there and even so far inland as Wrocław. By about 1200 Silesian nuns were sending small boats down the Odra to pick up “salt fish.”Footnote ⁸⁴ An intensified fishery off the island of Rügen, however, seems a twelfth-century development, soon contested between Slavic and Danish lords.Footnote ⁸⁵

Parallel developments along continental coasts of the North Sea and Channel from Flanders to Normandy are signaled in cartloads of herring on the market at Arras in 1024, an annual herring fair at Fecamp by 1088, and herring remains at eleventh-century sites as far inland as Compiègne, Paris, and Namur, though in the latter only at the castle, not yet the less affluent town. By the second half of the twelfth century communities along the Flemish coast had gained a papal dispensation to fish on Sundays while the run was on, and were struggling bitterly with old neighbouring monasteries over payment of tithes on their catch, which assumed a large presence on their newly established fish markets. The material results of these efforts are plain in abundant herring bones found in twelfth- and thirteenth-century urban contexts (see Figure 2.2 above).Footnote ⁸⁶

Across the narrow seas historian Maryanne Kowaleski estimates a four-fold increase in English herring catches from c. 1000 to c. 1200. Domesday Book (1085–1087) counted well over a score of large site-specific renders in herrings owed the king and East Anglian lords. This argues for a widespread fishery, if seasonal and by part-timers, and accounts for the newly high proportion of herring remains in food waste from twelfth-century Norwich, London, and other locations in eastern England. Transport some distance inland is revealed by mentions of herring in toll schedules.Footnote ⁸⁷ Exploitation also of local herring stocks in southwestern England sustained payments of 30,000 herring a year from Tidenham (where the Wye enters the Severn estuary) to the minster of Bath. When herring came to 60 percent of the fish bones gnawed by the monks of Norman Eynsham, did those rations come from the Severn or the North Sea?Footnote ⁸⁸

Most everywhere at this time the greater share of this catch was piled up whole right on the foreshore and covered with salt, making loose dry ‘powdered’ (sapoudre) herring with a few months’ storage life, then sold in bundles or baskets of a thousand to consumers as much as a hundred kilometers away.Footnote ⁸⁹ Already in the 1160s Alan of Lille called herring “the most common of fish, [which] by his wide availability relieved the hunger of the poor.”Footnote ⁹⁰ Plainly the surging exploitation of abundant herring stocks along Christendom’s northwestern littoral during the eleventh and twelfth centuries contributed heavily to increased quantities of marine fishes serving to meet the growing demand for fish as food in maritime Europe.

At least into the 1200s, however, written and archaeozoological records known to date indicate that consumption of herring remained an essentially regional phenomenon which rapidly attenuated towards the continental interior. Even in the twelfth century herring bones in Parisian trash still number but two-thirds those of eel. Further inland eleventh- and twelfth-century written evidence is confined to ecclesiastical settings while material finds are absent even from well-sieved sites.Footnote ⁹¹ In the eleventh-century Rhine basin monastic writers at Lorsch and St. Gallen describe the fish in learned terms referring to Roman fish pickle, not dry salted objects, and the only physical traces come from a castle near Basel. By the mid-1100s verbal acquaintance among monks at Hirsau and of Hildegard of Bingen with the preserved product gains but tepid confirmation from rare finds in slightly later urban latrines at Basel.Footnote ⁹² In Bavaria no herring bones are reported in a survey of fish remains from six castles and four urban contexts predating 1200Footnote ⁹³ nor do they leave any trace in either the Iberian or Italian peninsulas. Further east in the Baltic than Viking-Age Birka in Sweden, herring likewise occur only after 1200.Footnote ⁹⁴

Expanding eel fisheries at (western) European scale met growing medieval demand for fish by exploiting a species whose habitat itself was then growing in unintended consequence of human activities. Herring, however, epitomize intensified human use in northwestern coastal areas of an existing fish stock under conditions where larger cultural and demographic demand pressures confronted limited and probably dwindling supply from hitherto preferred varieties taken in fresh water. In a third emerging case during the same tenth through thirteenth centuries, the westward spread and human use of common carp replicated some features of the two previous, but reached across the extensive upper Danube basin into neighbouring waters then little affected by intensified consumption of eel or herring.

5.2.3 Exotic Carp Invade the West

The tale of carp in central medieval times relates the progress of what would now be called an exotic and invasive species. Throughout Greco-Roman antiquity common carp resided in European waters only in the Black Sea drainage, where remains from some Balkan sites suggest it was abundant. The northwesternmost traces of this fish up to and throughout the Roman Empire placed it in the Vienna basin, the most westerly part of Pannonia.Footnote ⁹⁵ Not coincidentally, this natural post-Pleistocene range of a slow-water, heat-loving but otherwise broadly tolerant, species ended precisely where it encountered the fast-moving, high-gradient waters of the upper Danube above that river’s ‘inland delta’ in the vicinity of Bratislava. Contrary to assertions by some biologists, no written or material evidence suggests that Romans dealt with carp outside that native range or there handled carp any differently than they did other freshwater fishes.Footnote ⁹⁶

In what remains the oldest known verbal European reference to this organism, in the mid-530s Cassiodorus, the learned Roman minister for the Gothic kings who ruled early sixth-century Italy, listed carpam destinet Danubius (“the Danube sends carp”) among the several exotic fishes the king would serve to impress visiting ambassadors.Footnote ⁹⁷ From Cassiodorus’ northern Italian standpoint, the pertinent Danube lay to the east in Pannonia (modern Hungary or Serbia), whence the Goths had entered Italy a generation before, where they then still also ruled, and where the carp were not only native but long consumed by humans. Tellingly, the only known material evidence of this fish from medieval Italy is isolated fragments from sixth–seventh-century Comacchio and Padua,Footnote ⁹⁸ precisely the time and route for such a luxury import to be transported to the king’s palace.

During the half-millennium from Cassiodorus into the eleventh century, finds of carp remains continue to confirm its natural distribution, and for the first time indicate its spread (1) further up the Danube and (2) north into middle reaches of the Odra and Elbe systems (Map 5.1). No verbal sources corroborate the northward expansion (presumably via low divides at the headwaters of the Morava River), perhaps because literate Latin Christian culture penetrated this zone only around 1000 CE and initially produced none of the requisite records of economic activity. Carp’s early medieval westward push, however, did leave traces in the eleventh-century written record. About 1060 the anonymous author at Tegernsee abbey of a secular fairy tale (courtly novel), composed in Latin, listed charpho together with other fishes, some named in Latin, others in the vernacular but all otherwise familiar in upper Bavaria, which his hero, a knight named Ruodlieb, captures by quasi-magical means.Footnote ⁹⁹ Barely a generation later, Abbot William of Hirsau (d.1091) compiled a vocabulary of signs for his monks to use during compulsory silence. To his prototype from Burgundian Cluny, William added several central European taxa, among them “the fish which is popularly called carp” (piscis qui vulgari nomine carpho dicitur). Did William then expect carp at Hirsau’s Black Forest site in the Rhine basin, or did he recall eating this fish during his youth at Regensburg on the Danube? Might William signify carp’s movement across watersheds?Footnote ¹⁰⁰ If so, this is corroborated by recent finds of carp remains from tenth-/early eleventh-century layers at Sulzbach castle, seat of the Count of Nordgau, in a zone of interlaced Danube and Rhine tributaries and then, barely a decade after William’s death, by a bilingual lexicographer at Lorsch abbey in the northern Black Forest who glossed a fish called in Latin carabus with the German charpho.Footnote ¹⁰¹ Carp remains of late eleventh- and early twelfth-century date at Nürnberg castle and a house of canons at Saarbrücken further fit this scenario.Footnote ¹⁰²

Map 5.1 The expanding range of common carp in Europe, 600–1600.

Five centuries from Bratislava to the Rhine; less than two from there to Paris? Mention by Hildegard of Bingen in her Physica (c. 1160) confirms the establishment of carp (carpo) in the middle Rhine. The abbess likely learned from abbey fishers how the species fed on bottom organisms and vegetation in swampy and clear water and something of its spawning habits, while herself assessing its value as food and, suitably prepared, a cure for fever.Footnote ¹⁰³ No remains or verbal mentions from before 1200 suggest carp culture or artificial fishponds. These were wild fish.

Further west the written and archaeozoological records seem to lag,Footnote ¹⁰⁴ and then suddenly blossom in the mid-thirteenth century. Does this mark arrival or literate recognition? A Parisian connection seems important. All three preeminent mid-century scholastic natural philosophers took cognizance of carp. Thomas of Cantimpre, who composed his Liber de natura rerum during the early 1240s, associated the fish with ponds and slow rivers. He had some idea of its morphology, fecundity, and ability to evade the fishers’ nets, but was ill-informed of reproductive behaviour. A decade or so later Vincent of Beauvais mostly replicated Thomas. Thus when writing De animalibus in 1258–1262, Albertus Magnus could select from and correct his predecessors. Perhaps drawing on his own experiences in Regensburg and Köln, Albert revised errors about spawning behaviour and commented on both the carp’s suitability for rearing in artificial ponds and its culinary qualities (which Albert doubted).Footnote ¹⁰⁵ Also in 1258 managers of the estate at Igny-le-Jard belonging to Thibaut VI, Count of Champagne, “stocked 3520 carpis and six big pike” costing more than eighty-three livres into one of their ponds. Later that same year they spent still more to put into two ponds some unidentified ‘fish’ (piscibus), 10,000 bream and roach (bremarum et gardonum), and 400 carpis, the latter alone costing nineteen livres.Footnote ¹⁰⁶ Igny-le-Jard, still well endowed with ponds, sits midway between Epernay and Chateau-Thierry, about three kilometers from the Marne and less than a hundred upstream of Paris. Within the next decade royal provost Etienne Boileau included carpes and cuerpiaus among the fishes of the Seine and other fresh waters which artisan fishers and fishmongers sold to Parisians.Footnote ¹⁰⁷ And from a mid-thirteenth-century rubbish pit beside the castle of the Louvre come the oldest known carp remains in France, followed in a human generation by the same from castles at Laarne and Londerzeel in Flanders.Footnote ¹⁰⁸ The carp had made itself home in Europe’s west-flowing watersheds.

How had this occurred? Not by entirely natural means, more as an unintended consequence of human activities. Although scholastic natural historians and Parisian regulations of the mid-thirteenth century treated carp as a wild fish, human agency with varying purposes in mind had for the previous half-millennium meshed tidily with the tolerances and tenacity of an aggressive organism to encourage its spread. As earlier here discussed regarding eel, during the sixth–eleventh centuries and later Europeans caught, prepared, and ate fish from their natural local waters. This often entailed live storage of seasonal catches in tanks, cages, or ponds close to such centers of consumption as elite residences, castles, or monasteries. Charlemagne had mandated this practice for royal estates in his capitulary De villis (c. 795), and other records suggest such vivaria were reasonably common.Footnote ¹⁰⁹

Ensuing centuries witnessed increasingly purposeful construction of individual ponds throughout western Christendom, some meant to keep fish (vivaria, servatoriae, piscinae), others functioning as mill ponds or moats but capable of holding fish, too. Lay seigneurs – as at eleventh-century Lanzenkirchen castle in Lower Austria – may have led this activity but churchmen like John of Metz, abbot of Gorze in Lorraine, accepted lay gifts of ponds and built more themselves. Radiocarbon dating to the period around 1000 of the oldest surviving artificial ponds in Berri matches their entry into the written record and local acceleration of woodland clearances.Footnote ¹¹⁰ By around 1200 – so well before carp are detected west of the Rhine – some ponds in Poitou, Maine, southern England, the Ile-de-France, Champagne, and Lorraine were equipped with adjustable sluices and bypass channels to manage their drainage and simplify mass harvest of fish.Footnote ¹¹¹ Even beyond the general effects of woodland clearance, watermills, and localized eutrophication, long before the carp arrived in the west, more or less unawares Europeans were preparing just the kind of habitat this fish would enjoy.

Captured wild fish stocked early medieval store ponds, whether those were isolated vivaria, run-of-the-river impoundments behind mill dams, or had other primary purposes. Nothing indicates any purposeful choice of variety or special care in storage; no document inventories these fishes by name. The result was inadvertent selection for tolerance of captivity, transport, and possible variations in water temperature, oxygen supply, and food. Only tough and resilient species survived, even when moved from one watershed to another, where a few escapees might colonize new territory on their own. This practice long continued for single or even multiple ponds where small wild fish were stocked for future growth, as is first explicitly described on the Count of Champagne’s estate at Provins in 1217–1219 (without naming the varieties) and in the 1230s in England on the bishop of Winchester’s estates with bream, perch, roach, eel, and pike.Footnote ¹¹² But just about that time French managers were beginning to put juveniles of single named species into chosen ponds, as seen in the carp at Igny-le-Jard.

Chapter 7 will explore implications of those ponds and carp domestication. For now the point is to acknowledge how synergies among rising medieval demand for fresh fish, multi-purpose human management of watercourses, and the resilient adaptability of carp jointly enabled that species to colonize continental Europe’s western watersheds by the mid-1200s.

5.4.1 Climatic and Hydrographic Fluctuations at Multiple Scales

Historical climatologists commonly describe the times treated in this chapter in terms of the ‘Medieval Climate Anomaly’ (henceforth MCA), a period in global climate history with patterns distinct from a most recent reference period (roughly 1880s–1980s/2000 or shorter) and from those of the intervening ‘Little Ice Age’ (henceforth LIA) (Figure 5.3). At global scale the MCA entailed a slightly warmed planet during the ninth through eleventh/twelfth centuries with peak temperatures c. 800–1000 and gradual cooling thereafter, “albeit with important differences regarding the timing and spatial extent” of these phenomena. Twenty-first-century climate historians stress that global or hemispheric averages are less relevant than the regional conditions wherein humans and other organisms experience the impact of climatic anomalies.Footnote ¹¹⁶

Figure 5.3 European mean summer temperature anomalies, 850–1550.

Energy flowing from the sun (solar irradiance) and terrestrial volcanism principally drove the earth’s premodern climate, with regional features a result of global oceanic and atmospheric circulation.Footnote ¹¹⁷ The MCA coincided with high solar activity, especially during 1080–1280, and flow of energy to the earth. Only one solar minimum during 1040–1080 (the Oort) took place during the MCA, which came to an end with the Wolf minimum, 1280–1350. The ensuing LIA, a period of cooler global climates, coincided with three more minima in rapid succession.Footnote ¹¹⁸ Volcanic eruptions introduce aerosols and dust into the atmosphere. These reduce arrival of solar energy to the Earth’s surface and often have a global or hemispheric cooling effect. After an intense cluster of large eruptions and cooling during the sixth century, few volcanoes affected the northern hemisphere until a series of eruptions during the 1150s through 1260s, followed by another in the 1340s. During much of the MCA global circulation patterns produced a positive phase in the North Atlantic Oscillation (henceforth NAO), meaning westerly flows from the Atlantic protected most of western Europe from colder air out of Siberia. From 1100 to 1260 Europe’s average annual temperatures surpassed twentieth-century norms. Growing seasons remained at or above twentieth-century means from the mid-tenth century through the 1250s, followed by a severe drop in the 1260s. That cool spell dissipated after 1270 when temperatures rebounded for about another human generation, followed, however, by great instability during middle decades of the fourteenth century.Footnote ¹¹⁹

Planetary or even large-scale regional average temperatures are only a part, and not necessarily all that important a part, of how climate and weather patterns affect living things.Footnote ¹²⁰ Fluctuations of hydroclimate (precipitation and evaporation) do not necessarily coincide with temperature in their timing nor the scale of variation. Specific regional manifestations of both temperature and precipitation differ from the large scale. Seasonal variations have the greatest impact at critical life-cycle stages for both humans and fishes. Extreme events and other environmental perturbations stress natural systems and societies dependent upon them.Footnote ¹²¹ So possible connections between climate and past fisheries need to work at smaller scale.

Neither the writer and readers of this book nor their medieval predecessors have direct physical perceptions of the abstraction ‘climate’ nor of its changes. Without serial records and the patterns those may reveal, humans experience meteorological events (weather), sometimes later recall extremes (cold, storms, heat, drought), and may over time adjust their behaviour to patterned experience. When does the snow come? How big a bridge will survive floods? When to expect the wheat harvest or the bream to assemble in shallows to spawn? Catastrophic events can transform habitats; gradual shifts in water temperature or chemistry can have similar effect, though perceived by humans only as one kind of fish replacing another. And over time human practices may change – though not due to recognition of ‘climate change’.

If the intensely studied Tiber watershed in central Italy can stand for conditions in western Mediterranean Christendom, after cooler than average and increasingly wet weather from the fifth through the ninth centuries, a warming trend set in and rose to a peak between the mid eleventh century and mid twelfth only to cool again slowly into the fourteenth. The region was distinctly drier between about 1050 and 1350 than before or after; the Tiber rarely reached flood stage. Obstructed drainage, however, turned the valley floor around Rieti into a permanent wetland of slow-moving waters.Footnote ¹²² Surely the environment for Carolingian-age fisheries of Farfa abbey (Chapter 3, pp. 101 and 107 above) had evolved. More general surveys of Italian conditions see changing phases, with c. 1100–1270 typically warm and arid, while lake core samples from eastern and coastal Spain give a similar impression.Footnote ¹²³ Subfossil remains of small vertebrate animals from sites along streams in Corsica show different patterns of diversity during droughty high medieval times than in subsequent wetter periods.Footnote ¹²⁴

North of the Alps the central European landscapes whence large rivers flow into the North Sea, Baltic, and Danube, experienced a different MCA. A period of peak temperatures was well defined in the mid-tenth to mid-eleventh centuries, followed by decadal or longer cold spells after 1050, in the early 1100s, and again in the early 1200s. Increased seasonality meant the thirteenth and early fourteenth centuries knew distinctly warm summers but chilly winters.Footnote ¹²⁵ The positive NAO also produced wetter summers, although tree ring data from both Scandinavia and the Swiss Alps signals somewhat drier conditions in those boundary areas.Footnote ¹²⁶ Overall warmer summer waters might inhibit successful reproduction by fishes intolerant of high temperatures while enabling successful spawning by species preferring those conditions.

On maritime fringes off northwestern Europe the warming trend of the MCA came early and may have reached its maximum sooner, too. Greenland was at its warmest c. 800–1000, reaching 1.8°C above the reference period. Norse settlers arrived at Iceland in 870 during a hundred-year warm spell, suffered a chilly eleventh century, but then between 1100 and the mid-1200s enjoyed the warmest summers in three centuries. After 1280 cooling set in across all seasons. In northwestern Scotland only a mid-eleventh-century drought interrupted otherwise wet conditions from around 900 to 1340. Sea surface temperatures in the North Atlantic – assessed from the barely visible fossil shells of temperature-specific diatoms in bottom cores – fluctuated at the decade scale through most of the twelfth century but then maintained high values from 1170 to 1260. A hundred years of cooling followed.Footnote ¹²⁷

Put simply, around 1100, when Mediterranean Europe was close to its warmest and aridity threatened some aquatic systems there, central Europe was already cooling down, especially in winter time, but quite well watered, and the North Atlantic coasts and islands, though very wet, retained a moderated climate, though less warm than it had been some centuries before. Organisms, ecosystems, and cultural adaptations at the edge of their critical tolerances had the most to lose or to gain from marginal changes.

Regional climate itself is but one component in the dynamic processes visible at medieval Europe’s land–water interfaces, its marine estuaries, coastal marshes, and open shorelines as well as the courses and banks of its rivers. To begin at the headwaters, many preindustrial rivers took an unstable local course. Seasonal patterns of flood, drought, and (in the north) ice cover kept channels mobile and banks impermanent. During high water rivers revisited their floodplains. Above the estuaries low water commonly revealed braided structures – bars, islands, shallows, multiple channels – as well as dominant constrictions, rapids, reefs, and waterfalls.Footnote ¹²⁸ To the historic interplay of climate, topography, soils, and human constructions rivers naturally responded with changes in morphology and their associated aquatic habitats.

Distinct periods of frequent and large regional inundations associated with short-term climatic instability occurred both during the MCA and as it drew to an end. Rivers in English lowlands showed rapid seasonal fluctuations and flood events during 1085–1117 and then again in the thirteenth century, which triggered structural changes into multi-channel forms.Footnote ¹²⁹ On the lower Rhine in the 1150s Emperor Frederick Barbarossa instructed the bishop of Utrecht and counts of Gelders, Kleve, and Holland to take measures to protect their subjects against flooding. Two centuries later and further up the Rhine two religious institutions went to canon law courts over changes to an island.Footnote ¹³⁰ Along both the upper Rhône and the rivers of the Pyrenees, after long docile centuries, frequent and disastrous flooding set in from the end of the twelfth century, rapidly increased through the thirteenth, and reached a peak during the 1350s–60s.Footnote ¹³¹ During that last wave of floods the upper Rhône transformed from a meandering to a braided morphology,Footnote ¹³² characterized by a higher gradient, coarser bed materials, large interannual and seasonal variability, and thus newly unstable habitats for aquatic life. Fluvial instability with human consequences even drew the mid-fourteenth-century attention of famed Italian jurist Bartolus of Sassoferrato and generations of late medieval Hungarian legists.Footnote ¹³³

Perturbations in riverine hydrologies flow down to estuaries and the sea. At the onset of the MCA in the tenth–eleventh centuries, upper and middle reaches of the Wisła became unusually active, with clusters of frequent flood events. In some areas a meandering morphology began to braid.Footnote ¹³⁴ At the time this large basin drained into a delta and estuary extending some sixty kilometers from Gdańsk east to Elbląg with bays, wetlands, and channels reaching equally far inland and all loosely delineated from the Baltic by a string of offshore islands. Archaeozoology reveals a brackish environment with inshore schooling fishes. From the thirteenth century and into the later Middle Ages this landscape evolved into an increasingly drained delta on the west, where the river entered the sea, and to the east a separate lagoon (Zalew Wislany, Frisches Haff) fully enclosed by a continuous sand spit.Footnote ¹³⁵ Fourteenth-century villagers along the lagoon shoreline there now took diadromous and freshwater fishes (salmon, pike), while by the mid-1400s sea-caught cod had become prominent in demesne fisheries of the ruling Teutonic Order.Footnote ¹³⁶

Coastal environments elsewhere experienced diverse variabilities. Average global sea level remained relatively stable throughout the mid–late Holocene, fluctuating only a meter or two. But local seismic activity, post-glacial uplift, land subsidence, shoreline erosion, terrestrial drainages, siltation, and shifts in storminess and normal wind direction variously affected medieval coastlines in several regions. Sweden’s Lake Mälaren was a bay of the Baltic open to seagoing vessels in the Viking Age and until about 1200, when post-glacial rebound raised it above mean sea level so it became entirely fresh water and Stockholm replaced Birka as the principal port. Salinity in the long, narrow estuary of the Schlei fell significantly in the course of the twelfth century, so what trace element analysis indicates were fishes taken locally ceased to be the marine species, herrings and others, eaten at Haithabu before 1085. Instead late medieval people barely five kilometers away at Schleswig ate perch, bream, and pike.Footnote ¹³⁷

Along the southern shore of the North Sea, storms could turn land into new arms of the sea. What had into the 1100s been a complex of freshwater lakes, wetlands, rivers, and drained farmland had become by the end of the 1200s a marine embayment, locally called the ‘South Sea’ (Zuiderzee). Land subsidence and storm floods broke the Frisian coastal barrier at Texel in 1282, and five years later the Saint Lucia flood of 14 December washed deep inland to drown tens of thousands of people and untold livestock, replacing thousands of hectares of pasture and arable with salty waters. While subject to intermittent storm surges, the new sea’s boundaries stabilized in the fifteenth century and lasted into the twentieth.Footnote ¹³⁸ Storm waves from a different direction washed deep into drained wetlands of the Thames estuary in the 1230s, 1280s, 1320s,1334, and 1370s; once people stopped trying restore the damage, the newly enlarged salt marsh increased local fish habitats and fisheries there expanded, especially after 1351.Footnote ¹³⁹ On the other hand, the Bay of the Somme which had reached to Abbeville twenty kilometers inland, was by 1300 receding together with its fisheries for eel and flatfishes, and so, too, further south the once even larger ‘Gulf of Pictons’ in Poitou. Yet on the other French coast, unstable passes and fluctuating salinity levels of Languedoc lagoons motivated fishers and rights holders to adjust their institutional relations and keep on fishing.Footnote ¹⁴⁰

Little recent or extant research on the history of climate or hydrology has paid direct attention to aquatic habitats. The norm in water history treats a physical fluid, not a biological substance. And indeed most any efforts to relate these variables had until recently to contend with climate data so broadly drawn (in both spatial and temporal terms) that conclusions would be crude at best and more often misguided. Even with increased palaeoscientific precision in assessing climatic variables, only very occasionally during the Middle Ages are coincident biological and/or economic archives of specific local fisheries sufficiently detailed to allow even probable hypotheses. But to assume the aquatic realm of medieval Europe was unchanging or driven only by human actions would fly in the face of basic ecological knowledge. While leaving problems of large commercial marine fisheries at the very end of the Middle Ages for treatment in Chapter 8 below, shifting ranges and stocks of several species on the continent and in the Baltic do appear consilient with climatic and hydrographic fluctuations.

5.4.2 Traces of Impacts, Resilience, and Adaptation

In the context of varying medieval climates and weather with evident or likely effects on regional and local aquatic systems, we can trace certain fairly well documented changes in specific fish stocks to prior or simultaneous natural phenomena. At least two freshwater fishes may have achieved their sometimes tenuous natural establishment in western Europe under favourable conditions of the MCA and then, lacking purposeful human intervention, barely survived the LIA there. Water temperature, salinity, and oxygen content influenced interacting species of considerable human interest in the medieval Baltic.

The rapid eleventh- through thirteenth-century spread already observed of common carp across central and western continental Europe was less evidently associated with people actually rearing fish – nearly all records before the 1250s treat carp as wild – than with their placing captured wild fish in ponds for live storage. This occurred during what is now understood as a warm phase of the MCA, when mean European annual and summer temperatures both peaked. Carp are thermophilic organisms, meaning they do like heat. In the wild this species begins to spawn in weedy shallows during May–June as water temperatures move above 18°C. Reproductive success is restricted to years when the water level starts rising in May and when high temperatures and flooding of terrestrial vegetation last for a long period during May and June. This is because carp larvae survive only in what is by modern north European standards very warm water (at or above 20°C) among shallow submerged vegetation.Footnote ¹⁴¹ Chapter 7 will show that later European fish farmers learned to design special spawning ponds to warm quickly and to handle the larvae with special care, but that came as the climate cooled into the LIA. Carp’s prior high medieval expansion likely benefitted as much from natural heat as it did from humans moving a prospective fish dinner from one watershed to another.

Adult individuals of a small cyprinid, the bitterling, look very like juvenile carp.Footnote ¹⁴² Until about 1100 the known range of the bitterling was restricted to southeastern Europe, more or less similar to the pre-medieval range of the carp, though no one ever thought bitterling would make a palatable meal. Bitterling are perhaps more thermophilic than carp, preferring waters above 16°C by June and needing 23°C for successful reproduction. The earliest records of bitterling in central and western Europe occur in regions where carp were becoming known and carp culture would later become significant (see Chapter 7). In the mid-twelfth century Hildegard of Bingen, where the Nahe joins the middle Rhine, even knew the bitterling’s love for warm water. Thereafter written, visual, and archaeozoological traces of bitterling multiply in those regions, only nearly to vanish after about 1550, while Europe endured the coldest two centuries of the LIA. Bitterling reappear and spread only from the late 1700s. While quite widely abundant in recent times, local populations decline markedly following years and decades with cold spring temperatures and revive with warmer ones as at the end of the twentieth century. Like carp, then, bitterling spread westwards during warm medieval summers with entirely inadvertent human assistance; come the cooler state of the LIA the species contracted most of its western range.Footnote ¹⁴³

Medieval evolution of herring and cod fisheries inside the Baltic may owe more to natural variability than to human enterprise or impacts. To recapitulate and anticipate: herring were abundant in the central and southwestern Baltic (Bornholm, Pomerania, the Schlei) from the fifth/sixth centuries into the thirteenth and by the latter date also in the Danish straits (of which much more in Chapter 8). Cod were certainly present in the Baltic during the Neolithic and again from the fourteenth century, while during early and high medieval times this species is virtually absent from the written and archaeozoological record there.Footnote ¹⁴⁴ The traditional explanation is that Slavic peoples had no taste for cod but immigrants from Germany did. Stable isotope and other studies of the cod bones themselves, however, identify the oldest medieval cod remains found at Haithabu and in the eastern Baltic as imports from the North Sea or Norway, that is, as imported stockfish. Yet by the end of the Middle Ages people in this region did plainly fish locally for cod.

More recent suggestions observe that known low salinity and hypoxic (low oxygen) conditions in deep water basins of the medieval Baltic could have suppressed cod populations. In the Baltic today strong hypoxic conditions thought to arise from a westerly flow of wind and waters from the Atlantic, a warming climate, and eutrophication from nutrient-rich runoff place a cap of warmed fresh water on top of colder saltier waters where no mixing occurs. This situation drives many free-floating eggs of cod into the deepest basins where lack of oxygen prevents larval development and so threatens cod recruitment.Footnote ¹⁴⁵ Baltic herring, however, like their North Sea kin, breed in relatively warm, biologically productive, and often brackish surface layers. Sediment cores demonstrate that hypoxia in the Baltic is not just a modern but rather a recurring phenomenon, present both during the early Holocene (c. 9000–c. 5000 ybp) and roughly between 550 CE and 1200 ± 50 years. Thereafter bottom waters became better oxygenated and remained so into the nineteenth century.Footnote ¹⁴⁶

Long-term changes in the state of the Baltic, termed a ‘regime shift’ by recent ecologists, have been linked to climate variations. The earlier medieval and the most recent condition of deep water hypoxia beneath a warm surface layer of low salinity result from a positive NAO characteristic of both the MCA and modern warming. Predominant westerlies then push salt water in from the North Sea and raise precipitation levels across the Baltic watersheds. Temperature and salinity differences encourage stratification. Near-surface temperatures and biological productivity are high. Fading of the MCA from around 1200 tended to bring more negative NAO, with lower salinity but also lower temperatures and nutrient levels. More balanced salinity and temperature allowed greater mixing, so conveying more oxygen to the deep basins. Distinctive onset of the LIA by around 1500 strengthened a negative NAO, produced drier and colder conditions in central European lands, and prolonged the no longer new marine trophic regime for another three centuries.

In the present-day Baltic abundant cod are the principal predator on herring, which controls the herring population. Under modern conditions of intensive fisheries, removal of predators results in superabundant prey populations, a typical form of trophic cascade. With few cod present, early medieval herring stocks could explode until limited by some other ecological factor.Footnote ¹⁴⁷ Very recent ecological field work and theories even suggest that, as herring will themselves predate on planktonic cod eggs and larvae, the large medieval schools could (further?) have suppressed a Baltic cod stock already under stress from low oxygen in the habitats critical for its reproductive life stage. Both the abundance of adult cod and cod recruitment show negative correlation with herring biomass.Footnote ¹⁴⁸

Based on organic and chemical markers in layered bottom sediments, the accepted chronology for shifts in oxygen and salinity levels in the medieval and early modern Baltic matches poorly with growth of human populations and the intensity of land use in that watershed. The initial conditions which would impact both cod and herring stocks were established well before high medieval clearances in central Europe and Scandinavia could much have affected the runoff regime. Indeed trends shift in the opposite direction just as the wave of settlement and clearances began to crest in the north German, Polish, and eastern river basins which provide the bulk of the sea’s fresh water. Although properties of large enclosed water bodies change more slowly than do atmospheric drivers, the southern Baltic herring fishery spiraled into insignificance in the course of the 1200s, supplanted by surging growth in the Danish straits. As Chapter 8 will trace closely, the latter fishery endured some two centuries during the very time fishing for cod came forward in many Baltic coastal areas. Those local cod fisheries would then persist even as Øresund herrings faded before Dutch exploitation of North Sea stocks under now stabilized LIA conditions. In sum, given what is now known about habitat requirements and predator–prey relationships across life cycles of cod and of herring, climatic and hydrological conditions in the Baltic Sea during the central and high Middle Ages (the MCA) were propitious for the latter species and stressful for the former. Climatic forces thus appear to have been important drivers in the medieval Baltic oscillation to and then from what one ecologist calls ‘a herring dominated state.’

Coastal societies exploited abundantly accessible stocks until natural conditions slowly changed. As foreshadowed earlier in this chapter and spotlighted in Chapter 8, the Baltic herring fishery of the tenth through fifteenth centuries was quite probably medieval Europe’s largest single source of marine protein.

And were Mediterranean waters and fishes unaffected? Have archaeologists and historians failed to ask the right questions in the most record-rich region of medieval Europe? Or did the natural world simply interest literate medieval Italian, Provençal, and Catalan elites even less than it did those of the north? Nevertheless, good proxies indicate colder Italian winters during the early 1300s and again in certain fifteenth-century decades. We also do know that eighteenth-century fisheries for Adriatic sardines fluctuated in tandem with weather patterns.Footnote ¹⁴⁹

~~~ <>< ~~~~ ><> ~~~

At local and regional scale aquatic systems of high medieval Europe were subject to multiple pressures and constraints across the interplay of natural and cultural forces. Relevant drivers and effects varied from one socio-natural site to another. A widely evident rise in human environmental impact had negative consequences for fishes with strict habitat requirements, while favouring varieties more tolerant of heat, low oxygen, and high nutrient levels. Changes of natural origin, a warmer climate and more, also created, destroyed, or shifted equilibria among fish species, with at least some effect on human use. Contemporary Europeans may have been oblivious to some such variations or lacked the perspective to see them in the longer term, but people plainly did become aware of certain changes. Some traditional fisheries seemed less productive (scarce) and other local stocks to offer fishers opportunities to respond to greater demand for fish. Neither resource destruction and depletion nor the dilemmas of allocation and conservation are peculiar to present-day fisheries crises. The next chapter turns to responses in medieval communities to perceptions of limits, declines, and shortages in fishes Europeans had long liked to eat, so exploring Europeans’ own cultural resilience and adaptability. Reciprocal and reiterative interactions among medieval European natures and cultures drove the larger narrative of fisheries to be apprehended as collectivities of myriad socio-natural sites.