September 03, 2009

Central European farmers not descended from local hunter-gatherers (Bramanti et al. 2009)

This is the real power of DNA: the topic of whether central European farmers were the result of demic diffusion from the southeast or indigenous hunter-gatherers who adopted the agricultural economy has been endlessly debated in archaeological circles.

We are finally in a position to give an answer to the question, and the answer is in favor of the diffusionist camp and against the idea of acculturation by local hunter-gatherers. Surprisingly, modern Central Europeans do not appear to be a simple hunter-gatherer/farmer mix, suggesting that even later events (post-Neolithic) have shaped their genetic diversity.

This study is also a powerful argument against the idea of genetic continuity across long time spans. Most ancient DNA studies so far have reached a similar conclusion. Thus, it also destroys the supposed justification for continuity from Paleolithic Europe to modern times that early mtDNA work (of the Daughters of Eve variety) has proposed, hand in hand with the hunter acculturation hypothesis.

The paper is covered in National Geographic:
Central and western Europe's first farmers weren't crafty, native hunter-gatherers who gradually gave up their spears for seeds, a new study says.

Instead, they were experienced outsiders who arrived on the scene around 5500 B.C. with animals in tow—and the locals apparently didn't roll out the welcome wagon.

"Within a few generations, all the farmers—probably coming from southeast Europe—moved into central Europe bringing their culture, [livestock], and everything," Joachim Burger, a molecular archaeologist at the University of Mainz in Germany, said via email.

The finding is based on analysis of genetic material in the skeletal remains of ancient hunter-gatherers and early farmers found in Germany, Lithuania, Poland, and Russia—though farming is thought to have reached areas as far west as western France during the period of rapid expansion, about 7,500 years ago.

The study goes against a long-standing idea that Europe's first farmers were former hunter-gatherer populations that had settled the region after the last ice age, about 10,000 years ago.

Perhaps, the thinking went, the hunter-gatherers had observed farming practices during their travels or had learned from neighbors.

Instead, the researchers found, the hunter-gatherers and the early farmers remained segregated, according to the study, to be published tomorrow in the journal Science.
And the press release:
Analysis of ancient DNA from skeletons suggests that Europe's first farmers were not the descendants of the people who settled the area after the retreat of the ice sheets. Instead, the early farmers probably migrated into major areas of central and eastern Europe about 7,500 years ago, bringing domesticated plants and animals with them, says Barbara Bramanti from Mainz University in Germany and colleagues. The researchers analyzed DNA from hunter-gatherer and early farmer burials, and compared those to each other and to the DNA of modern Europeans. They conclude that there is little evidence of a direct genetic link between the hunter-gatherers and the early farmers, and 82 percent of the types of mtDNA found in the hunter-gatherers are relatively rare in central Europeans today.

For more than a century archaeologists, anthropologists, linguists, and more recently, geneticists, have argued about who the ancestors of Europeans living today were. We know that people lived in Europe before and after the last big ice age and managed to survive by hunting and gathering. We also know that farming spread into Europe from the Near East over the last 9,000 years, thereby increasing the amount of food that can be produced by as much as 100-fold. But the extent to which modern Europeans are descended from either of those two groups has eluded scientists despite many attempts to answer this question.

Now, a team from Mainz University in Germany, together with researchers from UCL (University College London) and Cambridge, have found that the first farmers in central and northern Europe could not have been the descendents of the hunter-gatherers that came before them. But what is even more surprising, they also found that modern Europeans couldn't solely be the descendents of either the hunter-gatherer alone, or the first farmers alone, and are unlikely to be a mixture of just those two groups. "This is really odd", said Professor Mark Thomas, a population geneticist at UCL and co-author of the study. "For more than a century the debate has centered around how much we are the descendents of European hunter-gatherers and how much we are the descendents of Europe's early farmers. For the first time we are now able to directly compare the genes of these Stone Age Europeans, and what we find is that some DNA types just aren't there - despite being common in Europeans today."

Humans arrived in Europe 45,000 years ago and replaced the Neandertals. From that period on, European hunter-gatherers experienced lots of climatic changes, including the last Ice Age. After the end of the Ice Age, some 11,000 years ago, the hunter-gatherer lifestyle survived for a couple of thousand years but was then gradually replaced by agriculture. The question was whether this change in lifestyle from hunter-gatherer to farmer was brought to Europe by new people, or whether only the idea of farming spread. The new results from the Mainz-led team seems to solve much of this long standing debate.

"Our analysis shows that there is no direct continuity between hunter-gatherers and farmers in Central Europe," says Prof Joachim Burger. "As the hunter-gatherers were there first, the farmers must have immigrated into the area."

The study identifies the Carpathian Basin as the origin for early Central European farmers. "It seems that farmers of the Linearbandkeramik culture immigrated from what is modern day Hungary around 7,500 years ago into Central Europe, initially without mixing with local hunter gatherers," says Barbara Bramanti, first author of the study. "This is surprising, because there were cultural contacts between the locals and the immigrants, but, it appears, no genetic exchange of women."

The new study confirms what Joachim Burger´s team showed in 2005; that the first farmers were not the direct ancestors of modern European. Burger says "We are still searching for those remaining components of modern European ancestry. European hunter-gatherers and early farmers alone are not enough. But new ancient DNA data from later periods in European prehistory may shed also light on this in the future."
And from archaeology.about.com:
A new study published by Barbara Bramanti and colleagues in Science Express on September 4, 2009, supports what some scholars have suspected all along—that the LBK likely were an in-migration of people from the Balkans, and that they did not, initially anyway, do much mixing at all with the earlier inhabitants of Europe.

Bramanti and her colleagues compared the mitochondrial DNA from 20 central European Upper Paleolithic, Mesolithic and Neolithic hunter-gatherers to that from 25 Neolithic farmers and 484 modern Europeans, spanning an age range from about 13,400 to 2,300 BC. The data shows that the early farmers and hunter-gatherers were from distinctively different populations.

This paper follows up on and to a degree contradicts with the hypothesis of an earlier paper that looked only at mtDA of the Neolithic farmers. That study (Haak et al. 2005) discovered that the farmers had a distinctive difference between the current residents of Europe, and hypothesized that that meant that the hunter-gatherers might have been more like the modern inhabitants, and thus, the LBK would have been only a minor component.
The earlier paper by Haak et al. they refer to.

(More technical details once I read the full paper)

UPDATE:

Pre-farming populations seem to have been dominated by mtDNA haplogroup U:
it is intriguing to note that 82% of our 22 hunter-gatherer individuals carried clade U (fourteen U5, two U4, and two unspecified U-types; table 1).
The hunter-gatherers had no N1a -which was a signature of early farmers in the Haak et al. paper- or of haplogroup H, the most common mtDNA haplogroup in Europeans today. The only non-U types in hunter-gatherers were all from the Ostorf site and included haplogroups T2e, J, and K.

The farmers:
In a previous study, we showed that the early farmers of Central Europe carried mainly N1a, but also H, HV, J, K, T, V, and U3 types (11, 12). We found no U5 or U4 types in that early farmer sample.
UPDATE I:

It is important to note the implications of this study: the most certain conclusion is that Neolithic farmers in Central Europe are very sharply differentiated from the Paleolithic-Mesolithic populations. This is clear evidence in favor of the diffusionist idea, since the acculturation hypothesis predicts that the mtDNA of the early farmers would be roughly that of the pre-farming population that picked up the new technology.

However, the evidence of this paper also contradicts the plain demic diffusion hypothesis. According to this hypothesis, farmer genes are gradually replaced by hunter genes as the farming economy spreads, because in each step there is a mix of farmer-indigenous populations which go on to colonize regions beyond the frontier. This is not what appears to have happened. Rather, it seems the farmers moved across Europe with very little interaction with pre-farmers. A long period of no contact between the LBK and foragers is actually supported by archaeology. I have termed this type of diffusion the "skipping stone":
In the Skipping Stone model, farmers move out in search of new territories before they have started to blend with the local foragers; the genetic impact of the initiators of the movement is preserved.
The great speed of the Linearbandkeramik farmers was also experienced by farmers who spread across the Mediterranean. The spread of agriculture in Europe does not appear to have been a slow process of interaction between farmer and forager, but rather a blitz by the first farmers, followed later, after the spread had already occurred by admixture with some of the foragers that remained.

We must also be certain not to jump into conclusions about the relative contributions of farmer and forager in the modern gene pool. Clearly both the idea of a predominantly "Paleolithic" and a predominantly "Neolithic" gene pool is problematic; such continuity is not really evident. However, the reasons for the discontinuity up to the present may be manifold: e.g., later population movements into Europe, or natural selection changing the gene pool without subsequent change of population.

What we do know is this: first farmers were not local foragers who abandoned the old ways for the new ones. Amalgamation between farmer and forager did not happen quickly as the farming economy spread. Finally it did happen, of course, and either because (i) there were few foragers in the mix, or (ii) their mtDNA was selected against, modern central Europeans have very little mitochondrial descent from the earliest European populations.

PS: Natural selection against forager mtDNA is not very outlandish. For example, a severe reduction of U5a1 and U5b haplogroup in Britain from ancient to modern times has been observed, which could potentially mark another data point in a process of selection against that haplogroup over time.

My personal guess is that both demography and selection may have played a role in the marginalization of hunter-gatherer mtDNA . LBK farmers were already 3 thousand years removed from the earliest agriculturalists of the Near East, so it is conceivable that they had evolved an mtDNA gene pool adapted to the new lifestyle that outcompeted the indigenous European one. But, the long period of isolation from foragers may mean that only farmer mtDNA benefited from the demographic boom associated with the new economy, and by the time relations between the two groups warmed up, the relatively few newcomers already dwarfed the older population demographically.

UPDATE II (Sep 4):

To understand the magnitude of the difference between farmers and hunter-gatherers, the authors calculate their Fst=0.163, which can be compared with a maximum value of 0.0327 among modern Europeans and 0.133 for modern Eurasians from Europe to Australia. Subsequently, the authors test the hypotheses of (a) continuity between hunter-gatherers and farmers, and (b) continuity between hunter-gatherers and modern Central Europeans, rejecting both.

This isn't very surprising in the light of the anthropological evidence in favor of diffusion of farmers from the Near East and against the acculturation hypothesis presented recently by Pinhasi et al. The very close relationship of the LBK skulls and their proximity to samples from Nea Nikomedeia in Greece and Catal Hoyuk in Anatolia contrasts with the Mesolithic populations.

UPDATE III (Sep 21):

Some possible anthropological evidence for post-LBK infusion into Central Europe:
Mesolithic Europeans display considerable variation in humero-clavicular and brachial indices yet none approach the extreme "hyper-polar" morphology of LBK humans from the MESV. In contrast, Late Neolithic and Early Bronze Age peoples display elongated brachial and crural indices reminiscent of terminal Pleistocene and "tropically adapted" recent humans. These marked morphological changes likely reflect exogenous immigration during the terminal Fourth millennium cal BC.

Science doi:10.1126/science.1176869

Genetic Discontinuity Between Local Hunter-Gatherers and Central Europe’s First Farmers

B. Bramanti et al.

Following the domestication of animals and crops in the Near East some 11,000 years ago, farming reached much of Central Europe by 7,500 years before present. The extent to which these early European farmers were immigrants, or descendants of resident hunter-gatherers who had adopted farming, has been widely debated. We compare new mitochondrial DNA (mtDNA) sequences from late European hunter-gatherer skeletons with those from early farmers, and from modern Europeans. We find large genetic differences between all three groups that cannot be explained by population continuity alone. Most (82%) of the ancient hunter-gatherers share mtDNA types that are relatively rare in Central Europeans today. Together, these analyses provide persuasive evidence that the first farmers were not the descendants of local hunter-gatherers but immigrated into Central Europe at the onset of the Neolithic.

Link

21 comments:

Gioiello said...

It seems to me that this paper demonstrates above all one thing: that mtDNA K (mine), born probably from U8b in North Italy, is present among the hunter-gatherers, among the Neolithic farmers and among today’s European population (and not only). That YDNA R1b1b2 couldn’t have had the same destiny I’ll believe when a similar study will be done on YDNA.
Interestingly we don’t find, among the numerous U5b subclades, the Italian U5b3, that expanded surely to South, certainly with mtDNA K we find till Middle East and, if possible, YDNA R1b1b2.

eurologist said...

MtDNA is fickle and any conclusions must be taken with a grain of salt. Even today, I could show you small regions within Europe with predominantly a single MtDNA sub-sub-group. That's just how local regions and their inhabitants work...

Clearly, there is nothing spectacular about the results. The farmer's MtDNA is largely of the same type that is widely spread within Europe, today (50% H, 11% J, 10% T, sizable K and T).

So what, N1a is not predominant (while widely present), now. It happened to be accidentally overrepresented in a few families then - simply not meaningful. Pretty much all of the found "agricultural" MtDNA, with very little exception, is old (pre-LGM) European.


Similarly, the observed "hunter-gatherer" MtDNA - as expected - only provides a restricted snapshot of the known distributions - which are all not only old, but still make up 11% (U5) and 10% (T) of modern Europeans.

Given the many chance developments and small population groups, it is outright laughable how anyone could argue for discontinuity versus continuity , based on these data.

Dienekes said...

Even today, I could show you small regions within Europe with predominantly a single MtDNA sub-sub-group. That's just how local regions and their inhabitants work...

The pre-Neolithic samples are not from a "local region" but include individuals from Germany to Russia and from 13,400 calBC to 2,250 calBC calBC. Non-U mtDNA is found only in the Ostorf site, which is both one of the latest ones and one that was an enclave of Mesolithic hunters surrounded by the Funnel Beaker culture.

So, yes, the evidence for haplogroup U in pre-Neolithic populations is probably as good as it can get.

but still make up 11% (U5) and 10% (T) of modern Europeans.

T is not necessarily pre-Neolithic:

"We are cautious of interpreting this as a signature of
local admixture (17), particularly as the hunter-gatherer and
early farmer T2 types belong to different sublineages, but it is
notable that Ostorf is culturally a Mesolithic enclave
surrounded by Neolithic funnel beaker farmers and is the only
hunter-gatherer site where any non-U mtDNA types were
observed (Table 1)."

As for U5 its frequency is:

"Europeans today have moderate frequencies of U5 types,
ranging from about 1-5% along the Mediterranean coastline
to 5-7% in most core European areas, and rising to 10-20% in
northeastern European Uralic-speakers, with a maximum of
over 40% in the Scandinavian Saami."

eurologist said...

You know very well that arguing a few percentages over seven millennia is futile.

You have to look at the origin and major inhabited regions and opportunities for huge multiplication of local groups.

The widely advertised renditions of this paper are clearly overusing the available data and reasonable interpretation in every single aspect imaginable.

Dienekes said...

You know very well that arguing a few percentages over seven millennia is futile.

An ~8-fold reduction in the frequency of the haplogroups prevalent in hunter-gatherers is not "a few percentages". The modern frequencies are well known and the ancient ones are reasonably estimated from a pre-Neolithic sample that is varied both geographically and temporally.

Jake myers said...

It seems to me to be likely that the spread of these Farmers West/NW/SW into other parts of Europe from the Balkans & Eastern Europe is Directly Related to the permanent Great Flooding of the Black Sea Lowlands, which were likely Farming lands, which occurred Circa 5550 BCE. Such a disruption likely relocated Flood refugees up the western Ri\/eR \/alleys into the Balkans, Carpathians, & Central Europe, pushing their nati\/es further west & this Unstable Migration Period likely lasted a few hundred years.

Creative said...

I can’t remember details, but a while ago I read an article on Neolithic violence and tensions between farming communities and hunter-gatherers in Central Europe. Documented by individuals that showed signs of force that lead to death.

The National Museum of Denmark in Copenhagen has a skull on display demonstrating late Neolithic violence.I once read elsewhere that this skull documentates violent tensions between Farming communities and hunter-gatherers.

Porsmose,Skull
http://oldtiden.natmus.dk/udstillingen/bondestenalderen/slebne_oekser_af_flint/vaaben_vold_og_doed_i_bondestenalderen/language/uk/

Jake myers said...

Great notice on this skull find.It makes a lot of sense. It also reminds me of all the ancient Akkadian, Sumerian & Sanskrit stories about the conflicts between the Farmers (likely Lowlander & \/alley dwellers) & The Shepherds (likely Highlanders & Mountain dwellers), the most famous being the Cain & able story. Shepherds were much more likely to be Hunter-Gatherers than were the farmers, but I do think it quite likely Sumerian & Sanskit speaking Highlanders from south of the Black Sea, all the way east to Kashmir, were likely both Farmers , Shepherds, Hunter Gatherer's & originators of Bronze `Age Metallurgy, \/iniculture, Trade cara\/ans, & big users of wheels, oxen, horses, baskets & pottery. They may be the leaders, whom led the migrations west from the Black Sea. some scholars say \/inca has the Earliest Finds of Written scripts.

eurologist said...

Dienekes,

What I mean is that the ancient hunter-gatherer MtDNA has survived quite well - both the observed one (which excludes smaller percentage haplogroups because of the small sample size), and the inferred one (much of today's MtDNA in Europe is thought to be ancient European, present at least during post-LGM expansion, and likely 10,000 or more years before then). Sure, U* was reduced from 80% or something to 15% - 20%, those things happen, and it apparently was not farmer MtDNA that makes up for the difference, but mostly other old European MtDNA.

Likewise, we don't have the slightest idea where that N1a came from. It could well have been present at the same 0.2% level Eurasian-wide as it is now. Apparently, the early LBK farmer families had it. It could very well be chance, it could very, very well be just one of the local semi-settled hunter-gatherer++ haplogroups in that valley of the Danube where these already semi-settled folks (fishing, triticum, and flax while modern farmers came from Anatolia) figured out a way to make those foreign crops and animals grow and live in that much different climate...

For those reasons, what we can't deduce is this:

What we do know is this: first farmers were not local foragers who abandoned the old ways for the new ones.

or:

Europeans have very little mitochondrial descent from the earliest European populations.

Quite the contrary.

Dienekes said...

Sumerian & Sanskrit stories about the conflicts between the Farmers (likely Lowlander & \/alley dwellers) & The Shepherds (likely Highlanders & Mountain dwellers), the most famous being the Cain & able story. Shepherds were much more likely to be Hunter-Gatherers than were the farmers

The contrast between Farmers and Shepherds occurs within Neolithic society itself. Sheep (and other domesticated animals) arrived with farmers. Hunters hunted animals, they didn't herd them.

terryt said...

"Shepherds were much more likely to be Hunter-Gatherers than were the farmers".

We have three separate categories here. Farmers are no more likely to be hunter-gatherers than are shepherds. If anything farmers would have more down time to go hunting than would shepherds. Besides which I'm pretty sure the 'Flooding of the Black Sea Lowlands' took place long before 'Circa 5550 BCE'.

Gioiello said...

5600BCE: Brian fagan, The Long Summer.How Climate Changed Civilization. This theory of the farmers on the North shore of Black Sea who migrated to North giving life to the Linearbandkeramik peoples and modern Europeans is a good theory. They probably were R1b1b2 and spoke an Indo-European language. Having among them many mtDNA K, born in Italy, should make us think to a previous origin, at list in part, from Italy- North Balkans. I believe always to the hypothesis that Rhaeto-Etruscan-Pelasgians were linked to an ancient phase of Indo-European.

argiedude said...

Did the study count samples that were identical in haplotype and haplogroup and came from the same site as a single sample? Was there any criticism of the 2005 study that found 25% N1a?

This study does constitute pretty good evidence in favor of a population expansion instead of cultural transmission, but it hasn't settled the matter. Dienekes talks about an 8-fold reduction in U frequency, but there's a similar problem with the farmer mtdna. Their main haplogroup, N1a, has suffered a 25-fold reduction. And today the rate of H in Europe is 50%, while the study found it at only 16% in the farmers, so that's another big change, a 3-fold increase.

Gioiello said...

And also K has been halved, but farmer had multiplaied it for 4. But we are always playing at home.

Ponto said...

Glad I am mtDNA V. Its pedigree while disputed due to lack of finds in not so ancient inhabitants of Iberia, has a purely European origin. mtDNA K is somewhat foreign, but Oetzi's odd haplogroup shows that it was more diverse in the Iron Age.

The farming communities may have been violent. Where in the world now is violence not found? Some cases of people who had their heads bashed in, in Neolithic times do not make it the rule for all those previous inhabitants of Europe. In Malta there were no weapons found archeologically until after the coming of Bronze Age people and there are signs there that the island was inhabited by Neolithic peoples 7 Kya. Oetzi, the Alpine Iron Age man, himself died of the results of an arrow wound.

The Neolithic age farmers moved about Europe in a saltatory manner, forming colony like communities complete with wheat, barley, pigs, sheep, goats, cattle and their pottery. They moved over long distances by land or sea to places where agriculture would flourish. They naturally were surrounded by in situ Hunter/Gatherers wherever they went. You cannot rule out the predominance of Hunter/Gatherers genetically over Neolithic farmers just based on comparing the haplogroups modern Europeans ca 21st century with long dead Europeans whether they were Hunter/Gatherers or not. It has been shown from Neolithic age human remains that those people were not continuous as far as haplogroups go with modern European groups. So modern Europeans are neither like the Hunter/Gatherers according to the study, nor like Neolithic Europeans according to other studies. Europeans today have a unique set of haplogroups containing some of those of previous inhabitants of Europe but almost showing no flow on genetic relationship.

Gioiello said...

Ponto says: “Europeans today have a unique set of haplogroups containing some of those of previous inhabitants of Europe but almost showing no flow on genetic relationship”.

If we had found L, M or D in ancient Europeans, we could think that there is no flow, but founding H, U, K, etc., sometimes of a different line than today, we can think that we are always playing at home. Probably modern haplogroups continue some haplogroups that were rare and that have survived, and other then more diffused are extinct, but certainly we are the descendants of those Europeans.
K 224c/311c is mine, and probably K 224c/258g/311c has become extinct or has had a back mutation, what does it matter? We are always playing at home.
That Oetzi has lived here, in Italy, perhaps with a different haplotype now extinct, demonstrates always that K was born and flourished here and not elsewhere.

Gioiello said...

Re the other mutation present in the Ancient Europeans, 16249C, we have two entry in Mitosearch: GTF6R and VW7QM, and they are K1a like me.

Geneboy said...

I am not really suprised at all, this pattern of replacement or swamping is known from the nordic and finnic colonisation of saami areas where the saami behaived as peaceful retreaters in most cases as they where quickly outnumbered resulting in the modern saami settlement restricted to highland or remote areas with minimal genetic inprint in the earlier settlement areas, however of course it is possible that the saamis in these long gone areas had a different genetic profiles of the saamis known today.

terryt said...

"this pattern of replacement or swamping is known from the nordic and finnic colonisation of saami areas".

Not only. Perhaps the same process was involved in the replacement of Neanderthals.

eurologist said...

Pre-Columbian North America may be another good example. Farmers and hunter-gatherers lived side-by-side; sometimes peacefully, sometimes not. AFAIK there is no documentation that either group was above-average, particularly violent before the Europeans made them battle each other.

What may be important is that even during the LBK "Blitz," things would have appeared somewhat slow on a generational time frame. The hunter-gatherers saw a couple more settlements in their region during their lifetime, of people who seemed to compete little for the wild life, but brought with them the opportunity to barter (say, grains, ceramic vessels, string/rope, clothing, hand-crafted vanity items vs. stones, axes, furs, collected berries and nuts, other, different hand-crafted vanity items on the other side, and also brought with them exotic, pretty women.

On the flip side, the MtDNA seems to support theories that claim early farmers were (at least partially) matriarchal, with farming women very protective of their farming knowledge, rights, standing, and their being the sole source of children.

It may have been a strong taboo for a farmer's boy to bring home a "dirty, poor, uneducated tom-boy girl from the forest." If they eloped, her MtDNA would have remained with the hunter-gatherers.

Also, at least during the first several generations of expansion, it is likely that the hunter-gatherers would have had the better weapons and better training/use of weapons. I don't see (initially) the aggressive spread of y-DNA that Europeans later documented throughout the world, and the related rapid inclusion of "native" MtDNA from multiple "illegitimate" children of mistresses. The farmers would not have been dominant like that, until they far outnumbered the hunter-gatherers (500 to 2,000 years later, depending on location).

Stay tuned for y-DNA and perhaps, one day, x-DNA analyses... ;)

Annie Mouse said...

I am not convinced by the conclusions of this paper. Wolfgang Haaks paper shows Haplogroup U, H and K living together in the same high density community in Germany during the late neolithic (4000-5000 years). H was in the minority.

We know that mobile hunter gathers had spaced children to cope with the lifestyle. A child had to be old enough to survive before the mother could have another child.

We also know that the early sedentary farmers had no such constraints and bred like rabbits.

It has been suggested that there is a genetic and/or dietary component to this difference.

One possible explanation for the high percentage of Haplogroup H in Europe is that this Haplogroup is simply better able to expand rapidly when the opportunity arises.