Founder

PHYLOGENY

Analysis in 2011 determined that GD (genetic distance) between the STRs of Rox2 matches formed a smooth bell curve.  This indicated that Rox2 was probably a monophyletic clade descended from one man.  That conclusion gained credence with subsequent NGS/SNP analysis.  Three Rox2 kits took 'Next Generation Sequencing' FGC tests in 2014.  Those early pioneers were the first Rox2 matches to get their results included on the Big Tree, thus helping define the size of the phylogenetically equivalent SNP block shared by all Rox2 matches.  The FGC SNPs are also present in the more recent higher-resolution Big Y-700 tests (not BigY-500).  There was no change in the early phylogeny after Big Y-700 testing.  Underneath Rox2's phylogenetically equivalent block a number of more recent SNPs, known at FTDNA's Block Tree as 'Private Variants', have accumulated over time for members of each parallel 'brother' subclade down to the present day.

A similarly shallow phylogenetic structure would occur if an eighteenth century man, perhaps one from a distant land with no other living male relations, had eight sons in Britain who also went on to have lots of children of their own.  If his many living descendants took a yDNA test they would find that they all shared SNPs in a large block leading from prehistory to the founder's time of birth (c. 1750).  There would be few unique Private Variants in evidence in their Big Y tests.  That man's descendants would all be very close ySTR matches to each other - the average genetic distance (GD) between the cluster's present-day STR haplotypes would be low.  If, however, the same haplogroup founder was born four hundred years earlier in 1350 AD, then over half a dozen reliable Private Variant SNPs could have occurred in that time.  The STR haplotypes would, on average, have a higher GD too.  Rox2 was born c. 700 AD (plus or minus a wide margin of error), and that is evidenced in the larger number of Private Variants and higher GDs present in the results of the eight parallel branches.

Tracking individual yDNA lines might be imagined like following a long thread of cotton - yDNA is a small part of our overall genetic makeup but it is a real and tangible (albeit thin) line connecting back to a distant ancestor.  In Rox2's case the thread leads back through more than forty father-to-son generations and a maze of over thirteen centuries.  A founder appears to have been active in northern Europe in the Early Middle Ages.  Of course, the Rox2 yDNA lineage existed before then but the apparent 'bottleneck' reveals that only one patriarch produced a large number of descendants.  Time to most recent common ancestor (TMRCA) estimates suggest that was about 1300 years ago (c. 700 AD).  Subclades downstream of Rox2 appear to be the result of an Early Medieval Period dynastic founding event in northern Europe.  Many of the 'sons' look to have produced equally large founding events of their own.

We probably will never know the circumstances involved in the founding of Rox2.  Genetic bottlenecks can result from the migration of an individual from 'country A' followed by a founding event (many sons) some distance away, in 'country B'.  If the founding event was significant enough to produce many surviving male offspring the resulting yDNA cluster is visible in modern hobbyist databases.  Rox2's eight known 'sons' are indicative of a much larger founding event, one that perhaps involved more than one partner and resulted in the births of at least as many daughters. The long list of SNPs in the shared block and the average number of unique family SNPs after the shared block give an indication of a clade's potential time of arrival in country B.  Bottlenecks can occur if the yDNA lineage is 'pruned' to only one male who goes on to have many surviving offspring.  Such a scenario might result from a chance natural extinction in which all male yDNA relations in a population group - sons, uncles and cousins and very distant cousins - diminish to just one surviving male.  A man-made conflict or accident could also conceivably result in the almost complete loss of the male members of an extended family.

An event like the 'worst year to be alive' of 536 AD might produce such a bottleneck if an extended family/tribe was living in a remote place that was severely affected by adverse climactic conditions.  The latest ancient ice sample analysis indicates that this sixth century crisis was caused by the eruption of an Icelandic volcano.  Scandinavia was particularly badly affected - that time was known as Fimbulwinter.  The ability to grow crops at northern latitudes would have been hampered and may have prompted people to move - it may have even been the catalyst that hastened the collapse of the Roman Empire and increased the momentum of the Migration Period and the Scandinavian diaspora during the Viking Age.  The agricultural archaeological evidence described by Gräslund and Price shows that almost 75% of villages were abandoned in parts of Sweden and 'areas of southern Norway show a decrease in formal burials—indicating that haste was required in interments—up to 90-95%.' (link).  The Vendel Period in Scandinavia began in about 540 AD.  Just after this mini-ice age the Plague of Justinian that started in 541 AD and recurred in waves over the next couple of centuries further reduced the population in Europe.

In another loose analogy Rox2 might be imagined in gardening terms as a long and thin tree trunk that suddenly bursts into eight branches with a very bushy canopy.  It is possible that older geographically and genealogically distant 'cousins' have not been picked up yet in the mainly Ireland and British Isles-heavy American hobbyist databases.  If the source population that produced the founder died out, that location might never be found unless, by extreme good fortune, archaeological remains are found and analysed.  Population numbers were comparatively low 1000+ years ago - if Rox2 came from a small family in an isolated or marginal region there might now be no trace of them.

UNTANGLING THE CANOPY

There are well over one hundred different surnames at 111 STR marker resolution and above that are represented in the cluster.  Breaks in the connection between the yDNA line and the ancestral surname can happen, known in DNA genealogy as a 'NPE', or non-paternity event.  NPEs are not always the result of illegitimacy and can occur for any number of reasons.  See this blog post by Maurice Gleeson on the subject, link.  Surnames are not usually a reliable way of tracing back over 1000 years.  NPEs that result in surname changes might have occurred at any time in the 900 years since surnames began to be adopted but the inherited yDNA branch structure stays the same.  

Geneticist Prof Turi King of Leicester University estimates a 1-2 % chance of a NPE occurring per generation - this figure might be higher or lower for individual subclades.  If the surname/branch is around 900 years old, then a 2% NPE rate per generation would result in a cumulative NPE rate of over 60% today - meaning that two thirds of the present-day surnames in the phylogenetic trees could be different to the surnames of the yDNA lineage founders.  Given the tiny number of the population who have taken yDNA tests for genealogy it is quite possible that all modern descendants of some subclades now bear different surnames to the medieval founders of those lineages.  However, surnames can still give helpful hints about the general geographic locations and origins of EKAs (Earliest Known Ancestors) in parish records, mostly beginning in around the sixteenth century.

The Industrial Revolution of the late-eighteenth and nineteenth century saw populations break free of their feudal and local parish ties, resulting in movement around the British Isles to the growing cities and overseas to British colonies.  Modern hobbyist DNA databases represent where some of a subclade's descendants were in relatively recent times.  The distribution of pre-Industrial Revolution, locally-specific surnames suggests that Rox2's descendants were probably mobile in the Early Medieval Period and had expanded quickly across Britain and Ireland - a wide demographic expansion that happened before the general adoption of surnames.  Rox2 families usually have surnames that are familiar in their home regions, be that in rural England, Ireland, Scotland, Isle of Man or Sweden.  Many surnames in Northern Ireland appear to have earlier Scottish roots while many of those in the Republic have Irish surnames.  Maps showing present day Rox2 surname distribution reflect emigration patterns in the last few hundred years.  Significantly more people in former British colonies around the world have taken yDNA tests for genealogy than those in the 'home' countries themselves.

SPECULATION

As with surname history research, it is important to look at the historical context for the relevant eras and geographies.  The STR haplotypes and NGS results of all Rox2 matches converge back to one point but the generous margin of error attached to all age estimates must be considered.  It is likely there was increased mixing of yDNA lineages in the maritime regions of northwest Europe from Late Bronze Age/Iron Age times onwards - La Tène Celts influenced a wide area of the northwest and mingled with British tribes, Belgic tribes and the Jastorf culture who, in turn, later migrated around the North Sea and settled in what is now Britain, Scandinavia, Holland, Belgium and northern France.  A 300 year-or-so margin before the current TMRCA estimate takes the time of Rox2's expansion back to the early fifth century and the end of Roman rule in Britain and the arrival of Angles, Saxons, Jutes and Franks in the Migration Period.  300 years after 700 AD came the Normans, Flemings and Bretons.  Ancient DNA from archaeological studies, with accurate chronological dates and burial context for those samples, could provide the more precise timing evidence required.

Nevertheless, if the midway-point of current age estimates for the birth of the Rox2 founder is correct and the distribution pattern on my map is not an artifact of bias in the hobbyist databases, then the rapid expansion of several parallel subclades across the north of Europe might suggest a temporal connection for some branches with the historically-attested westward diaspora of continental northern Europeans in the eighth and ninth centuries.  That temporal connection has no specific bearing on where the Rox2 founder originally lived or came from but might fit with the seemingly widespread distribution of the several contemporary descendant lineages.  The peoples whose way of life involved moving widely and rapidly in-and-around the British Isles, Isle of Man and Ireland (by boat) in the 700s-900s AD are known as vikings.  When written with a lower-case 'v' the term here refers to an occupation rather than to a distinct people.  The 'Viking World' covered a large maritime area and was inhabited by a wide range of peoples who interacted within it, as found in Population genomics of the Viking world, Ashot Margaryan et al., 2019 and The Anglo-Saxon migration and the formation of the early English gene pool, Joshcha Gretzinger et al., 2022 (see Medieval DF27 page).  If the founder was born over 300 years earlier, as suggested by FTDNA Discover dates, then a period of founding activity and movement occurring in the late-Iron Age, Roman Period, sub-Roman Period or Migration Period should be considered.  The era spanning the end of Roman rule until the beginning of the Viking Age saw a development of wealthy coastal emporia visited by Frisian traders who acted as middlemen between Britain and the Carolingian empire.  Northumbria was attacked and Lindisfarne was plundered in 793.  The raiders returned the following year but were shipwrecked and defeated. 

The recreated burial-ship at Sutton Hoo.  By Gernot Keller www.gernot-keller.com, link

When looking at the North Sea basin one might imagine what is now Belgium, Holland, Germany and Denmark on the east coast and Northumbria, East Anglia and Kent on the west as being like two halves of an ancient North Sea territory with Norway above them and Francia below.  Surface currents enabled sea travel between East Anglia on one side of the southern North Sea and the region of the Rhine Delta, and above it, on the other.  The North Sea was also connected by busy maritime trade routes that circulated clockwise and anticlockwise around its shores throughout history.  The end of Roman rule in Britain in the late-fourth century AD led to a gradual increase in Anglian foederati influence. The North Sea mercenaries and their chieftains had manned Roman camps in northern and eastern England and their descendants eventually wrested control from their Romano-British employers.  The extreme weather events of 535-536 AD and the long-lasting effects of the Justinian plague of 541 coincided with Brittonic tales of the Gododdin in the Hen Ogledd ('The Old North').  The Vendel Period began c. 540 AD.  Rox2's founding event after its millennia-long bottleneck may well have happened around this time (suggested by FTDNA Discover TMRCA estimates) plus or minus a generous margin of error.  If so, rather than being a Viking Age phenomenon (as suggested by early 111T estimates), the subclade's roots might conceivably track back to the mists of the bitter sub-Roman struggle for eastern Britain.

The later Viking Age (late-eighth-to-eleventh centuries AD) was probably partly fueled by ongoing conflict between mixed descendants of earlier Anglian and Brittonic clans who nursed ancient historical claims over kingdoms around the North Sea  littoral since the sub-Roman period.  In Scandinavian tradition all sons, both legitimate and illegitimate, were considered to be the heirs of any of their extended paternal family, a system that could occasionally lead to conflict.  The Scaldingi, an extended kin-group that claimed descent from a mythical founder named Scyld, probably came from the same milieu and region of northern Europe that Angles had vacated a few hundred years previously and that Gaulish/Belgic tribes had left 1000 years before them.  The scale of the potential for complex circulation and mixing of uniparental DNA markers around the North Sea basin by mobile maritime groups in the Early Middle Ages is clear.

Surnames only became fixed for most people several centuries after the founding of the Rox2 subclade and it is difficult to know what is coincidence and what isn't.  However, some surnames match those of families with origins in Flanders and Normandy:

The friendship between Robert de Brus and David FitzMalcolm (after 1124 King David I of Scotland), who was present in France with King Henry and was granted much of the Cotentin Peninsula, may have commenced at least as early as 1120, at Henry's Court.   Link

Several families, traditionally associated with the move north from the Norman Court of Henry I in the households of David I, are outlined in Scottish Hazard, Volume Two: The Flemish Heritage, Beryl Platts, 1990.  Modern yDNA distribution maps appear to show a 'hot spot' of DF27 in the area of Old Francia that became the County of Boulogne (896–1501), Picardy and Flanders.  The Busby et al. study of yDNA distribution in Europe also found a relatively high frequency of modern P312xL21,U152, most of which will be DF27, in the region of Paris (North Central France where it made up the highest proportion of R1b-P310 at 17.6% ) as it did in Northwest France (20.9%).

The formation of the Scoto-Northumbrian realm of King David I of Scotland in the early twelfth century involved a northward migration from Flanders and Normandy in the household of David's wife Queen Maud, Countess of Huntingdon.  Maud had Flemish family connections through her mother, Judith of Lens, as well as being descended from the Earls of Northumbria through her father, Waltheof.  In the decades either side of the Norman Conquest the Comté de Boulogne in Pas-de-Calais had ties with the east coast of Britain through trade in the lucrative herring fishery.  While some Rox2 branches appear to have potentially been present in the British Isles and Ireland since at least the Early Middle Ages with TMRCA estimates centering on the Migration Period, other branches potentially exhibit a second period of expansion in northern Britain after 1066 AD - a timing that coincides with the arrival of David I, Robert de Brus and their Romance-speaking Boulonnais, Frankish, Flemish, Breton or 'Norman' entourages in the early-twelfth century.

HISTORICAL yDNA

Ancient yDNA from historic remains are increasingly being analysed in archaeological projects, as seen in the following recent examples.  The University of Leicester's The King in the Car Park gained worldwide press attention in 2013.  The yDNA of the 530-year-old skeleton did not match the yDNA of modern descendants of the Plantagenets, there had been NPEs or 'non-paternity events in the past.  As mentioned, NPEs are to be expected in every ancient lineage.  Four of the five claimed modern descendants did match each other but they did not match Richard III's remains.  Among other projects, Prof Turi King, leader of the University of Leicester team, mentioned on BBC Radio 4's The Life Scientific (broadcast 9th July, 2019) that they have a toe bone under analysis that belonged to Robert the Bruce and that they have a male descendant to compare the results with.  It was mentioned that there are more studies underway at Leicester but they're unable to elaborate due to confidentiality agreements.  

In February 2022 genealogy researchers from the University of Strathclyde announced in the press that they may have identified the yDNA of Robert the Bruce, King of Scots from 1306 to 1329, by testing descendants of Robert Bruce of Clackmannan's sons Robert and Edward.  The SNP found was FTB15831.  According to the FTDNA Haplotree FTB15831 descends from the Late Neolithic/Early Bronze Age SNP DF27>ZZ12>Z46512/FTT1>FGC78762, as does Rox2.  The two yDNA lines diverge at that early point.  The full phylogeny of the purported de Brus yDNA is DF27>ZZ12>Z46512/FTT1>FGC78762>FT3917>Y37728>S7437>S6219>FT378466>FTB15831.  The parentage of the Bruces of Clackmannan (descended from Thomas Bruce in the mid- to late-thirteenth century) is not certain, however.  Corroboration from comparison with descendants from other branches and archaeological yDNA would confirm the findings.  Graham Holton, Principal Tutor on Strathclyde’s Genealogical Studies Postgraduate Programme adds, 'This discovery will also allow the comparison of these results with any Y-DNA which can be extracted from supposed remains of King Robert, and thus confirm the true identity.'

Elsewhere, the mixed bones in caskets at Winchester Cathedral have been analysed and aDNA should be forthcoming.  A scientific paper covering the April 2019 UK Channel 4 TV program about the Great Army base at Repton by Catrine L. Jarman et al. was due to be released soon after but has been delayed.  A big study of Viking Age remains, Population genomics of the Viking world, Ashot Margaryan et al., was released in July 2019.  The Genetic History of France, Aude Saint Pierre et al. was also released in July 2019 but did not include yDNA results.  There are rumours (September 2019) that an elite Mycenaean burial is R1b (xZ2103 ).  For discussion of ancient DF27 discoveries see Ancient DF27.