In Northern Mists - BestLightNovel.com
You’re reading novel In Northern Mists Volume I Part 16 online at BestLightNovel.com. Please use the follow button to get notification about the latest chapter next time when you visit BestLightNovel.com. Use F11 button to read novel in full-screen(PC only). Drop by anytime you want to read free – fast – latest novel. It’s great if you could leave a comment, share your opinion about the new chapters, new novel with others on the internet. We’ll do our best to bring you the finest, latest novel everyday. Enjoy
[Ill.u.s.tration: s.h.i.+pment of tribute. From the bronze doors from Babavat, a.s.syria (British Museum)]
[Ill.u.s.tration: Wars.h.i.+p of Ramses III., circa 1200 B.C.]
[Sidenote: The earliest boats of Northern Europe]
It is difficult to say how the Scandinavians at the outset arrived at their boats and s.h.i.+ps, such as we know them from the boats found at Nydam in Sleswick and the Viking s.h.i.+ps discovered in Norwegian burial-mounds.
They are of the same type that in Norway, in the districts of Sunnmor and Nordland, has persisted to our time, and they show a mastery both in their lines and in their workmans.h.i.+p that must have required a long period for its development. From the accounts of many contemporaries, as well as from archaeological finds, we know that even so late as the first and second centuries A.D. large canoes, made of dug-out tree trunks, were in common use on the north coast of Germany between the Elbe and the Rhine, and there can be no doubt that this was the original form of boat in the north and west of Central Europe. In England similar canoes made of the dug-out trunks of oaks have been found with a length of as much as forty-eight feet; they have also been found in Scotland, in Bremen and in Sleswick-Holstein (in many cases over thirty-eight feet long), with holes for oars. It is related of the Saxons north of the Elbe, who at an early period made piratical raids on coasts to the south of them, that they sailed in small boats made of wicker-work, with an oaken keel and covered with hides. Besides these they clearly had dug-out canoes; but in the third century A.D. it is recorded that they built s.h.i.+ps on the Roman model. The only people north of the Mediterranean of whom we know with certainty that they had their own well-developed methods of s.h.i.+pbuilding are, as already mentioned (p. 39), the Veneti at the mouth of the Loire, whose powerful and seaworthy s.h.i.+ps of oak are described by Caesar. That the Scandinavians should have derived their methods from them cannot be regarded as probable, unless it can be proved that the intervening peoples possessed something more than primitive canoes and coracles. We must therefore believe, either that the Scandinavians developed their methods of s.h.i.+pbuilding quite independently, or that they had communication with the Mediterranean by some other route than the sea. Now in many important features there is such a great resemblance between the Norwegian Viking s.h.i.+ps and pictorial representations of Greek s.h.i.+ps, and of even earlier Egyptian and a.s.syrian s.h.i.+ps, that it is difficult to avoid the conclusion that some connection must have existed. For instance, the resemblance between the strikingly lofty prows and stern-posts, sometimes bent back, with characteristic ornamentation, and animal heads, which are already to be found in Egyptian and a.s.syrian representations, cannot be explained offhand as coincidences occurring in types independently developed. They are decorations, and cannot have contributed to the seaworthiness of the boats or had any practical purpose, unless the animal heads were intended to frighten enemies (?). It is true that lofty and remarkable prows are to be found in boats from such a widely separated region of culture as Polynesia; but in the first place it is not impossible that here too there may be a distant connection with the Orient, and in the second, the Mediterranean and Scandinavian forms of s.h.i.+p are so characteristic, compared with those of other parts of the world, that we necessarily place them apart as belonging to a distinct sphere of culture. Another characteristic of these boats and s.h.i.+ps is the oars with rowlocks (open or closed), instead of paddles. The rudder of the Viking s.h.i.+p (see ill.u.s.trations, pp. 246, 247, 248, 250) is also in appearance and mode of use so remarkably like the Egyptian rudder of as early as circa 1600 B.C.
(see ill.u.s.trations, pp. 7, 23), and the Greek (p. 48),[225] that it is not easy to believe that this, together with all the other resemblances, were independent discoveries of the North. The square sail and mast of the Scandinavian boat also closely resemble those of Egyptian, Phnician, Greek and Italian s.h.i.+ps as depicted.
[Ill.u.s.tration: Stone from Stenkyrka in Gotland (ninth century)]
It may be supposed that the communication which originally produced these resemblances did not take place by the sea-route, round the coasts of western Europe, but overland between the Black Sea and the Baltic. It is thus possible that the Scandinavian type of boat first began to be developed in the closed waters of the Baltic. It is here too that the boats of the Scandinavians (Suiones) are first mentioned in literature by Tacitus, and it is here that the earliest known boats of Scandinavian type have been found; these are the three remarkable boats of about the third century A.D. which were discovered at Nydam, near Flensburg. The best preserved of them (p. 110) is of oak, about seventy-eight feet long, with fourteen oars on each side, and it carried a crew of about forty men. The boats terminated in exactly the same way fore and aft, agreeing with what Tacitus says of the boats of the Suiones; and they could be rowed in both directions. They had rowlocks with oar-grummets like those in use on the west coast and in the northern part of Norway. There is no indication of the boats having had masts and sails, which also agrees with Tacitus.
There can be no doubt that we have here the typical Scandinavian form of boat, with such fine lines and such excellent workmans.h.i.+p that it can only be due to an ancient culture the development of which had extended over many centuries.
From the Baltic this form of boat may have spread to Norway, where it gradually attained its greatest perfection; and it is worth remarking that in that very district where the Baltic type of boat derived from the south-east reached a coast with superior harbours, richer fisheries, and better opportunities for longer sea voyages, namely, in Bohuslen and Viken, we find also the greater number of rock-carvings with representations of s.h.i.+ps. It is moreover a question whether the very name of "Vikings" is not connected with this district, and did not originally mean men from Viken, Vikvaerings; as they were specially prominent, the name finally became a common designation for all Scandinavians, as had formerly been the case with the names Eruli, Saxons, Danes.[226] In the course of their voyages towards the south-west the Scandinavians may also have met very early with s.h.i.+ps from the Mediterranean, which, for instance, were engaged in the tin trade with the south of England, or may even have reached the amber coast, and thus fresh influence from the Mediterranean may have been added. When we see how in the fifth century roving Eruli reached as far as Italy in their s.h.i.+ps, this will not appear impossible; and if there is any contrivance that we should expect to show a certain community of character over a wide area, it is surely the s.h.i.+p or boat.
Tacitus says that the fleets of the Suiones consisted of row-boats without sails. It is difficult to contest the accuracy of so definite a statement, especially as it is supported by the Nydam find, and by the circ.u.mstance that the Anglo-Saxons appear to have crossed the sea to Britain in nothing but row-boats; but Tacitus is speaking of wars.h.i.+ps in particular, and it is impossible that sails should not have been known and used in Scandinavia, and especially in Norway, at that time. There are possibly indications of sails even in the rock-carvings (see the first example in ill.u.s.tration, p. 236), and in the ornaments on the knives of the Bronze Age (see ill.u.s.tration, p. 238). In the case of a people whose lot it was to live to so great an extent on and by the sea, it is scarcely to be supposed that any very long time should elapse before they thought of making use of the wind, even if they did not originally derive the invention of sails from the Mediterranean.
Just as the Phnicians and the Greeks had swift-sailing longs.h.i.+ps for war and piracy, and other, broader sailing-s.h.i.+ps for trade (see p. 48), so also did the Scandinavians gradually develop two kinds of craft: the swift longs.h.i.+ps, and the broader and heavier trading-vessels, called "bosses"
and "knars."
[Sidenote: s.h.i.+pbuilding in Norway]
But even if northern s.h.i.+pbuilding exhibits a connection with that of the Mediterranean, and thus was no more spontaneous in its growth than any other form of culture in the world, the type of s.h.i.+p produced by the Scandinavians was nevertheless undoubtedly superior to all that had preceded it, just as they themselves were incontestably the most skilful seamen of their time. The perfection and refinement of form, with fine lines, which we find in the three preserved boats from Nydam, and in the three s.h.i.+ps of the beginning of the Viking age, or about the year 800, give evidence in each case of centuries of culture in this province; and when we see the richness of workmans.h.i.+p expended on the Oseberg s.h.i.+p and all the utensils that were found with it, we understand that it was no upstart race that produced all this, but a people that may well have sailed the North Sea even a thousand years earlier, in the time of Pytheas.
[Ill.u.s.tration: The preserved portion of the Viking s.h.i.+p from Gokstad, near Sandefjord (ninth century)]
The immigration to Norway of many tribes may itself have taken place by sea. Thus the Horder and Ryger are certainly the same tribes as the "Harudes" (the "Charudes" of the emperor Augustus and of Ptolemy), dwelling in Jutland and on the Rhine (cf. Caesar), and the "Rugii" west of the Vistula on the south coast of the Baltic (from whom possibly Rugen takes its name).[227] They came by the sea route to western Norway straight from Jutland and North Germany, and there must thus have been communication between these countries at that time; but how early we do not know; it may have been at the beginning of our era, and it may have been earlier.[228] But the fact that whole tribes were able to make so long a migration by sea indicates in any case a high development of navigation, and again it is on the Baltic that we first find it.
[Ill.u.s.tration: The Viking s.h.i.+p from Oseberg, near Tonsberg (ninth century)]
The s.h.i.+pbuilding and seamans.h.i.+p of the Norwegians mark a new epoch in the history both of navigation and discovery, and with their voyages the knowledge of northern lands and waters was at once completely changed. As previously pointed out (p. 170), we notice this change of period already in Ottar's communications to King Alfred, but their explorations of land and sea begin more particularly with the colonisation of Iceland, which in its turn became the starting-point for expeditions farther west.
We find accounts of these voyages of discovery in the old writings and sagas, a large part of which was put into writing in Iceland. A sombre undercurrent runs through these narratives of voyages in unknown seas; even though they may be partly legendary, they nevertheless bear witness in their terseness to the silent struggle of hardy men with ice, storms, cold and want, in the light summer and long, dark winter of the North.
[Ill.u.s.tration: s.h.i.+ps from the Bayeux tapestry (eleventh century)]
[Sidenote: The Norwegians' appliances for navigation]
They had neither compa.s.s, nor astronomical instruments, nor any of the appliances of our time for finding their position at sea; they could only sail by the sun, moon and stars, and it seems incomprehensible how for days and weeks, when these were invisible, they were able to find their course through fog and bad weather; but they found it, and the open craft of the Norwegian Vikings, with their square sails, fared north and west over the whole ocean, from Novaya Zemlya and Spitsbergen to Greenland, Baffin's Bay, Newfoundland and North America, and over these lands and seas the Nors.e.m.e.n extended their dominion. It was not till five hundred years later that the s.h.i.+ps of other nations were to make their way to the same regions.
[Ill.u.s.tration: Landing of William the Conqueror's s.h.i.+ps in England. Bayeux tapestry (eleventh century)]
The lodestone, or compa.s.s, did not reach the Norwegians till the thirteenth century.[229] As to what means they had before that time for finding their course at sea, Norse-Icelandic literature contains extremely scanty statements. We see that to them, as to the Phnicians before them, the pole-star was the lodestar, and that they sometimes used birds--ravens--to find out the direction of land; but we also hear that when they met with fog or cloudy weather they drifted without knowing where they were, and sometimes went in the opposite direction to that they expected, as in Thorstein Ericson's attempt to make Wineland from Greenland, where they arrived off Iceland instead of off America. Even when after a long period of dull weather they saw the sun again, it could not help them to determine their direction at all accurately, unless they knew the approximate time of day; but their sense of time was certainly far keener than ours, which has been blunted by the use of clocks. Several accounts show that on land the Scandinavians knew how to observe the sun accurately, in what quarter and at what time it set, how long the day or the night lasted at the summer or winter solstice, etc. From this they formed an idea of their northern lat.i.tude. Amongst other works a treatise of the close of the thirteenth century or later included in the fourth part of the collection "Rymbegla" [1780, pp. 472 ff.] shows that they may even have understood how to take primitive measurements of the sun's alt.i.tude at noon with a kind of quadrant. But they can scarcely have been able to take observations of this kind on board s.h.i.+p during their long voyages in early times, and they still less understood how to compute the lat.i.tude from such measurements except perhaps at the equinoxes and solstices. It is true that from the narrative, to be mentioned later, of a voyage in the north of Baffin's Bay, about 1267, it appears that at sea also they attempted to get an idea of the sun's alt.i.tude by observing where the shadow of the gunwale, on the side nearest the sun, fell on a man lying athwarts.h.i.+ps when the sun was in the south. With all its imperfection this shows that at least they observed the sun's alt.i.tude.[230] In order to form some idea of their western or eastern longitude they cannot have had any other means than reckoning; and so long as the sun and stars were visible, and they knew in what direction they were sailing, they undoubtedly had great skill in reckoning this. In thick weather they could still manage so long as the wind held unaltered; but they could not know when it changed; they were then obliged to judge from such signs as birds, of what country they were, and in what direction they flew; we hear occasionally that they had birds from Ireland, or from Iceland, and so on. The difference in the fauna of birds might give them information. In their sailing directions it is also stated that they observed the whale; thus in the Landnamabok (Hauksbok) we read that when sailing from Norway to Greenland one should keep far enough to the south of Iceland to have birds and whales from thence. This is more difficult to understand, as the whale is not confined to the land, and the same whales are found in various parts of the northern seas. But drift-ice or ice-bergs, if they met them, might serve to show their direction, as might occasionally driftwood or floating seaweed. The colour of the sea may certainly have been of importance to such keen observers, even though we hear nothing of it; it cannot have escaped them, for instance, that the water of the Gulf Stream was of a purer blue than the rather greenish-brown water of the coastal current near Norway and in the North Sea, or in the East Iceland Polar Current; the difference between the water of the East Greenland Polar Current and in the Atlantic is also striking. It may likewise be supposed that men who were dependent to such a degree on observing every sign may have remarked the distribution in the ocean of so striking a creature as the great red jelly-fish. If so, it may often have given them valuable information of their approximate position.
They used the lead, as appears, amongst other authorities, from the "Historia Norwegiae," where we read that Ingolf and Hjorleif found Iceland "by probing the waves with the lead."
[Ill.u.s.tration: Seal of the town of Dover, 1284]
But that it was not always easy to find their course is shown, amongst other instances, by the account of Eric the Red's settlement in Greenland, when twenty-five s.h.i.+ps left Iceland, but only fourteen are said to have arrived. Here, as elsewhere, it was the more capable commanders who came through.
[Ill.u.s.tration]
THE NORWEGIAN SETTLEMENT IN ICELAND
[Sidenote: Oldest authorities]
The island of Iceland is mentioned, as we have seen, for the first time in literature by Dicuil, in 825, who calls it Thyle and speaks of its discovery by the Irish. As he says nothing about "Nortmannic" pirates having arrived there, whereas he mentions their having expelled the Irish monks from the Faroes, we may conclude that the Nors.e.m.e.n had not yet reached Iceland at that time. The first certain mention of the name Iceland is in the German poem "Meregarto" (see p. 181),[231] and in Adam of Bremen, where we find the first description of the island derived from a Scandinavian source (see p. 193).
Narratives of its discovery by the Nors.e.m.e.n and of their first settlement there are to be found in Norse-Icelandic literature; but they were written down 250 or 300 years after the events. These narratives of the first discoverers mentioned by name and their deeds, which were handed down by tradition for so long a time, can therefore scarcely be regarded as more than legendary; nevertheless they may give us a picture in broad outlines of how voyages of discovery were accomplished in those times.
As the Norwegians visited the Scottish islands and Ireland many centuries before they discovered Iceland, it is highly probable that they had information from the Irish of this great island to the north-west; if so, it was natural that they should afterwards search for it, although according to most Norse-Icelandic accounts it is said to have been found accidentally by mariners driven out of their course.
[Sidenote: Are Frode on the settlement of Iceland]
According to the sagas a Norwegian Viking, Grim Kamban, had established himself in the Faroes (about 800 A.D.) and had expelled thence the Irish priests; but possibly there was a Celtic population, at any rate in the southern islands (cf. p. 164). After that time there was comparatively active communication between the islands and Norway, and it was on the way to the Faroes or to the Scottish islands that certain voyagers were said to have been driven northward by a storm to the great unknown island. The earliest and, without comparison, the most trustworthy authority, Are Frode,[232] gives in his "islendingabok" (of about 1120-1130) no information of any such discovery, and this fact does not tend to strengthen one's belief in it. Are tells us briefly and plainly:
"Iceland was first settled from Norway in the days of Harold Fairhair, the son of Halfdan the Black; it was at that time--according to Teit, Bishop Isleif's son, my foster-brother, the wisest man I have known, and Thorkel Gellisson, my uncle, whose memory was long, and Thorid, Snorre G.o.de's daughter, who was both exceeding wise and truthful--when Ivar, Ragnar Lodbrok's son, caused St. Edmund, the king of the Angles [i.e., the English king], to be slain. And that was 870 winters after the birth of Christ, as it is written in his saga. Ingolf hight the Norseman of whom it is truthfully related that he first fared thence [from Norway] to Iceland, when Harold Fairhair was sixteen winters old, and for the second time a few winters later; he settled south in Reykjarvik; the place is called Ingolfshovde; Minthakseyre, where he first came to land, but Ingolfsfell, west of olfoss, of which he afterwards possessed himself. At that time Iceland was clothed with forest [i.e., birch forest] from the mountains to the strand. There were Christian men here, whom the Nors.e.m.e.n called 'Papar' ..." and who were Irish, as already mentioned, pp. 165 f. "And then there was great resort of men hither from Norway, until King Harold forbade it, since he thought that the land [i.e., Norway] would be deserted," etc.
We may certainly a.s.sume that this description of Are's is at least as trustworthy as the later statements on the same subject; but as Are probably also wrote a larger islendingabok, which is now lost, there is a possibility that he there related the discovery of Iceland in greater detail, and that the later authors have drawn from it.
[Ill.u.s.tration: Dragon-s.h.i.+p with a king and warrior (from the Flateyjarbok, circa 1390)]
[Sidenote: Tjodrik Monk on the discovery of Iceland]
The next written account of the discovery of Iceland is found in the "Historia de Antiquitate regum Norwagiensium"[233] of the Norwegian monk Tjodrik (written about 1180), where we read:
"In Harold's ninth year--some think in his tenth--certain merchants sailed to the islands which we call 'Phariae' ['Faereyjar' == the Faroes]; there they were attacked by tempest and wearied long and sore, until at last they were driven by the sea to a far distant land, which some think to have been the island of Thule; but I cannot either confirm or deny this, as I do not know the true state of the matter.
They landed and wandered far and wide; but although they climbed mountains, they nowhere found trace of human habitation. When they returned to Norway they told of the country they had found and by their praises incited many to seek it. Among them especially a chief named Ingolf, from the district that is called Hordaland; he made ready a s.h.i.+p, a.s.sociated with himself his brother-in-law Hjorleif and many others, and sought and found the country we speak of, and began to settle it together with his companions, about the tenth year of Harold's reign. This was the beginning of the settlement of that country which we now call Iceland--unless we take into account that certain persons, very few in number, from Ireland (that is, little Britain) are believed to have been there in older times, to judge from certain books and other articles that were found after them.
Nevertheless two others preceded Ingolf in this matter; the first was named Garar, after whom the land was at first called Gararsholmr, the second was named Floki. But what I have related may suffice concerning this matter."
[Sidenote: "Historia Norwegiae"]
It is probable that Tjodrik Monk was acquainted with Are Frode's islendingabok, or at least had sources connected with it. In the "Historia Norwegiae" by an unknown Norwegian author (written according to G. Storm about 1180-1190, but probably later, in the thirteenth century)[234] we read of the discovery of Iceland [Storm, 1880, p. 92]:
"Next, to the west, comes the great island which by the Italians is called Ultima Tile; but now it is inhabited by a considerable mult.i.tude, while formerly it was waste land, and unknown to men, until the time of Harold Fairhair. Then certain Nors.e.m.e.n, namely Ingolf and Hjorleif, fled thither from their native land, being guilty of homicide, with their wives and children, and resorted to this island, which was first discovered by Gardar and afterwards by another (?), and found it at last, by probing the waves with the lead."
[Sidenote: The Landnama on the discovery of Iceland]
In Sturla's Landnamabok, called the Sturlubok, of about 1250, we find almost the same story of the first discovery as in Tjodrik Monk. It runs:
"Thus it is related that men were to go from Norway to the Faroes--some mention Naddodd the Viking among them--but were driven westward in the ocean and there found a great land. They went up a high mountain in the East-fjords and looked around them, whether they could see smoke or any sign that the land was inhabited, and they saw nothing. They returned in the autumn to the Faroes. And as they sailed from the land, much snow fell upon the mountains, and therefore they called the land Snowland. They praised the land much. It is now called Reydarfjeld in the East-fjords, where they landed, so said the priest Saemund the Learned. There was a man named Gardar Svavarsson, of Swedish kin, and he went forth to seek Snowland, by the advice of his mother, who had second sight. He reached land east of East Horn, where there was then a harbour. Gardar sailed around the country and proved that it was an island. He wintered in the north at Husavik in Skialfanda and there built a house. In the spring, when he was ready for sea, a man in a boat, whose name was Nattfari, was driven away from him, and a thrall and a bondwoman. He afterwards dwelt at the place called Natfaravik. Gardar then went to Norway and praised the land much. He was the father of Uni, the father of Hroar TunguG.o.di.
After that the land was called Gardarsholm, and there was then forest between the mountains and the strand."
In Hauk's Landnamabok (of the beginning of the fourteenth century) Gardar's voyage is mentioned as the first, and Naddodd's as the second, and it is said of Gardar that he was "son of Svavar the Swede; he owned lands in Sealand, but was born in Sweden. He went to the Southern isles [Hebrides] to fetch her father's inheritance for his wife. But as he was sailing through Pettlands firth [Pentland, between Orkney and Shetland] a storm drove him back, and he drifted westward in the ocean, etc." The Sturlubok was doubtless written some fifty years before Hauk's Landnamabok, and was the authority for the latter and for the lost Landnamabok of Styrmir enn froi[235] (ob. 1245); but as the copy that has come down to us of the Sturlubok is later (about 1400), many have thought that on this point the Hauksbok is more to be relied upon, and have therefore held that according to the oldest Icelandic tradition the Swedish-born Dane Gardar was the first Scandinavian discoverer of Iceland. Support for this view has also been found in the fact that in another pa.s.sage of the Sturlubok we read: "Uni, son of Gardar who first found Iceland." It has therefore been held that it was not till after 1300 that a transposition was made in the order of Gardar's and Naddodd's voyages at the beginning of the book [cf. F. Jonsson, 1900, p. x.x.x.]. But this a.s.sertion may be doubtful; it seems rather as though the Icelandic tradition itself was uncertain on this point. We have seen above that the Norwegian work "Historia Norwegiae" mentions Gardar as the first; while the yet earlier Tjodrik Monk [1177-1180] has a tale of a first accidental voyage to Iceland, which is the same, in parts word for word, as the stories of both the Sturlubok and the Hauksbok of Naddodd's voyage, only that Tjodrik mentions no name in connection with it. He certainly says later that Gardar and Floki went there before Ingolf; but this must mean that all three came after the first-mentioned nameless voyage. If we compare with this the vague expression of the Sturlubok that "some mention Naddodd the Viking" in connection with that first accidental voyage, the logical conclusion must be that there was an old tradition that some one, whose ident.i.ty is uncertain, had been long ago driven by weather to this Snowland, in the same way as there was a tradition in Iceland that Gunnbjorn had been driven long ago to Gunnbjornskerries, before Greenland was discovered by Eric the Red.
Some have then connected this first storm-driven mariner with a Norwegian Viking-name, Naddodd. Thus are legends formed. But the first man to circ.u.mnavigate the country and to become more closely acquainted with it was, according to the tradition, Gardar, whose name was more certainly known; for which reason he was also readily named as the first discoverer of the country (just as Eric the Red and not Gunnbjorn was named as the discoverer of Greenland). Hauk Erlendsson then, in agreement with this, amended the Landnamabok by placing Gardar's voyage first, while at the same time he made the mention of Naddodd more precise, which was necessary, since his was to be a later and therefore equally well-known voyage. He also gives Naddodd's kin, which is not alluded to in the Sturlubok. This hypothesis is strengthened by the latter's vague expression, above referred to, about Naddodd, and by the fact that only Gardar's and Floki's names are mentioned by Tjodrik Monk, and only Gardar and another (Floki ?) in the "Historia Norwegiae." If Naddodd's voyage had come after Gardar's, and consequently was equally well known, it would be strange that it should not be mentioned together with his and with the third voyage that succeeded them. But the whole question is of little importance, since, as we have said, these narratives must be regarded as mere legends.
The third voyage, according to both the Hauksbok and Sturlubok, was made by a great Viking named Floki Vilgerdarson. He fitted out in Rogaland to seek Gardarsholm (or Snowland). He took with him three ravens which