Mohan Pant， Shuji Funo： The Journey of Indus Grid in Town Planning From Mohenjodaro to Kathmandu Valley –A Study on the Patterns of Blocks and Plot Divisions in the Planning of Mohenjodaro (Taxila， Pakistan) and Thimi(Kathmandu Valley)， Proceedings of the 5th International Symposium on Architectural Interchange in Asia， AIJ， “Global Environment and Diversity of Asian Architecture”， June 1ー4， 2004， Matsue， Japan.
Mohenjodaro is the city representative of Indus civilization thatis said to have flourished around five thousand years ago, while Sirkap, also known as Taxila, was a reknowned cultural centre at the period from 5•h century BC to 2nd century AD. Thimi is still an existing town settlement within Kathmandu Valley. The three cities, seperated by enormous time span and_geographic space, exhibit certain features that is striking to draw the attention _oaf rchaeologists and the urban historians. All the three settlements employ exactly the same grid dimension in making the division of the urban blocks. The standard modules are derived from one of the standard measures of a hasta and its multiples--the'danda'and'rajju', which ar stated in·A,t-hasastra, the work of Kautilya, Prime Minister of Chandragupta of Maurya dynasty who ruled from Pataliputra (Patna, Bihar of India) during the later quarter of 41h century BC. This paper makes a morphological analysis of the revealed plans of Mohenjodaro, Taxila and Thimi, and demonstrates a relationship in their town plan, the division of quarter blocks and the plot divisions as shown by the built clusters and_street boundaries. This is the first direct evidence to link the urban civilization of Indus with the living settlements that continue to exist up to modem times.
Introduction
Mohenjodaro is a city in the Indus Valley plain tho ght to
have been flourished.around 2500 BC (Fig. 1, 2). This site of chalcolithic period was first �iscovered in 1922 by R. D. Banerjee1,and subsequently extensive excavation works followed under John Marshall, the then director.General of Archaeological Survey of India. Marshall
published the results of the work in three volumes in 1931.2 The various
sections that have been excavated in the
area revealed a dense
settlement with its main streets straight and buildings aligned with streets and built in a regular pattern. In addition, the drainage system and sanitary layouts of the city showed that the settlement of Mohenjodaro-was laid in a
planned manner
and the people did
have a fairly developed urban
culture. The regularity of the layout was in contrast to those found
irr the settlements of Mesopotamia.3 Ma_rshall, from 1913, earlier to his involvement in Mohenjcidaro, had begun the excavation
works on Taxila, an earlyhistoric city located at the east of the upper reaches of Indus River some 500 km northwest of Mohenj daro. Taxila incorporates several
settlements·among which Marshall excavated lower settlement known as Sirkap, and
the other one a
•
few hundred meters away at an elevated site. The upper strata of Sirkap revealed clearly that the town was a planned settlement with one main street and lanes laid out regularly and orthogonally to the main street. According to_Marshall, the city was settled from 2nd century BC following the·rule of Bactrean from 189 BC.4 Marshal, in his'Guide to Taxila'published in 1936 as ribes the planned settlement including the building of the city wall to Sakas who followed the Bactreans to rule the region of Gandhara. However, in his later works in the edited three volumes of Taxila published in 1951, he
ascribes the building of the planned city revealed
from
the Saka level to the Indo-Greeks (Bactreans).5 Although similar assumptions
made by Marshall with respect to other artifacts found
in this city has been refuted by other researchers--about which we will duely return in the later section of this paper, his suggestions on the
building of the town
in
Hellenistic principles is uncritically accepted by the
subsequent writers on Taxila.
Our attention to Taxila and to Mohenjodaro was drawn when the authors were seeking the possibility of
the existence of those planning features in the region that were discovered in an earlier study of Thimi, one of the ancient towns of Kathmandu Valley.6 1n this search, it was an incidental finding that Taxila used the similar
measurements in the division of its urban blocks as that of Thimi. Further excursion on the ancient settlements naturally led to Mohenjodaro where, surprisingly, it was found
that same modular dimensions were employed in the division of the settlement blocks as that found in Thimi and Sirkap.
The research works on Indus Valley civilization has come
longway in the past 60 years from the time of their first discovery. A great many settlements and sites have been discovered in the Sindu
and Punjab plains that belonged to this culture, which at the same t血e indicate various phases of its development. Settlements such as Kalibangan and Lothal further demonstrated that Indus people did have planned
town settlements.7 Metrological studies of Mainakar and Vij8 suggest the employment of standard system of weights and linear measure throughout the
Indus region. 1n spite of the extensiveness of Indus culture in the region and their developed urban culture, the knowledge on the town settlement of Sirkap until now is completely independent of the culture of Indus people. The three volumes on Sirkap and other works by Marshall
on Taxila don't mention Mohenjodaro or Harappa even once.9 It may be partly that no artifacts showing affinities with the Indus have been discovered in the excavated limits of Sirkap and partly because the study on the planning aspects of both Mohenjodaro and Taxila lag behind the comparative analytical works conducted on other individual artifacts such as earthen wares, coins, images, beads and individual buildings. However, Further, later excavation works within the region of Taxila, and at localities close to Sirkap itself has begun to show that the region was settled from almost the beginning periods of Indus civilization to as late as the first centuries of the first millenium
suggesting a probable continuity in the patterns of settlements in Taxila and Indus cities.1,u0 On the other
hand, connections of certain communities of Kathmandu Valley with the Indus people through migration have been
considered a possibility by some researchers although, at present, these are only
speculations. The present paper will, for the first time, make an analysis of the town plan of the three cities in terms of the division of the settlement blocks, building clusters, and the measurement standards employed in their planning and demonstrate a cultural affinity between the three cities. Although 1皿ited areas of excavation, which are often widely separated, present certain difficulties to make an analysis of the settlement plan, this study is particularly facilitated by the fact that the cluster of excavated areas in Mohenjodaro and Taxila at present provide the most extensive drawing documentation of the urban settlement
areas so far unearthed on the arhaeological explorations of prehistoric and early
historic period in South Asian region.
Method of Analysis
The study will analyse the layout of the streets, in particular, the dimensions of the blocks defined by the
building clusters
and the quarter
blocks that are revealed through the archaeological excavations in Mohenjodaro and Sirkap and the existing street patterns of Thimi. In the former two
settlements, it is possible to discern such cluster blocks which are formed by the exterior walls or the boundary walls of a group of adjoined
dwellings. These cluster blocks have well defined boundary lines, which either coincide with the street or lane, or are separated with the adjoing cluster block with a narrow gap which define their boundaries. It is also apparent that the residents of the neighbouring dweJiings
shared in
common the wall at the boundary. In such instances, the course of masonry of one of the neighbouring houses adjacent to the boundary wall
don't overlap and show a verticle joint between such walls. Another characteristic of such boundary lines of cluster blocks is that they show a
continuity through
several cluster
blocks, all the more emphasizing their nature
as boundaries
of plots. The next measurement will be along the settlement blocks defined by the interval of lanes or the streets. Thus the
side of the lane or the
boundaries of the blocks will define the terminal boundaries for the measurement and the dimensional analysis of the settlements.
The maps utilized for the purpose of this analysis of Mohenjodaro and Sirkap are both from Marshall
(1931 and 1951). The drawings are in 1:250, which gives
an accuracy considered
sufficient for the present purpose of analysis. The dimensions of Thimi are from
field
measurements
conducted
by the authors
in 1996.
A
preliminary study of the drawings showed that the layout of settlement blocks
was related to a modular dimension of 19.2 m. This dimension is a multiple of the grid discovered in the town of Thimi (38.5 m). Hence the grid of 19.2 m is superimposed to to all
these areas of Mohenjodaro to examine the extent of planned regularity of the settlement. Likewise, this same
grid is employed to study the block layouts of Sirkap. Thus, although the three cities are widely separated in space and
time, the analytical approach
will be the same to all the three cities. We will make their individual case study beginning from Mohenjodaro,
which, however,
is not intended to suggest a chronological sequence—a subject that remains outside the scope of this paper.
Mohenjodaro
The
excavated areas of Mohenjodaro are distributed in different locations as can be
seen in the
aerial photograph. These areas have been named as HR Area, VS心ea, DK心ea, Citadel
Area, and in like manner—the acronyines coming from the name of the
leader of the team of excavations. The
HR area
is defined by two streets running north-south. Within the region
of excavated area this
part lies at the
southern end. The wider street—Street 1, at the west is considered to be the Main Street of the city and is one of the major thoroughfare shown in the
reconstructed map of the grid of the road network
of the settlement by Piggot. The street at the west (Street 5) is narrower and its connection to the northern area is still unknown. This HR sector until now covers the widest area with high density of built clusters
among the excavated
areas of the city. VS Area at the north
of HR Area is
connected by the Main Street. The area was originally in two separate drawings, which for the sake of convenience and to test
the rigour of analysis are merged into one. The HR Area, VS Area and The DK area at the east of the Main
Street on which the detail drawingplans are available, and constitute large clusters among the excavated' areas, are selected for the purpose
of present analysis.
The Main Street and The Axis of Alignment
The dwelling clusters
or the blocks of Mohenjodaro, in general, follow the orthogonal pattern (Fig 3, 4, 5). But the alignment of the
building clusters and cluster groups in all the
areas show slight deviations
from
the
orthogonal system. For instance, the
HR section shows three different alignments in the development of the building clusters. Although the alignment of the Main Street within the range of this
area does not change, it can be observed that the alignment of the lanes within the quarters of the block
that run parallel to the Main Street have at certain sections gentle deviation from the orthogonal system.
The lanes are
also occasionaly found staggered. The building clusters at the interior following the alignment of the lanes
thus
create a number of alignments in the development pattern
of the settlement clusters. This situation creates an obvious difficulty to assume one reference axis for the superimposition
of the grid
as well as to proceed on the dimensional measurements that could span all the area. However, in this instance,
the
main street, which is straight throughout from south to the north of the city, is taken as the reference to proceed on the measurement work of the quarters and the cluster blocks.
These features apply in the same manner
to VS Area
at the north and and Dk area at the west.
The dimensional
measures obtained between the boundaries of the clusters,
between
the
lanes, and between the boundaries of the cluster
blocks and the
lanes of the Main
Area is
as shown in corresponding figures. It can be seen from the figure
that all the
lanes including the main street closely coincide and align
with a grid network superimposed over the settlement clusters which is drawn at the interval of 19.20 m. Further, the minor cul-de-sac lanes and the cluster boundaries are aligned with the
9.6 m grid. The
cluster blocks of 19.20 m width are most evident through the Dk and VS areas.
It is of importance to
note that despite the
four
different cluster
alignments, the boundaries in all cluster groups basically follow one reference
grid that aligns with the Main street. The various alignments are only a slight rotation from the
reference
grid which has resulted
not the offset but a bend showing the continuity in the grid lines. The conjunction of the grid lines of various cluster alignments with the main reference axis therefore
suggests their singular
origin. Such minor bendings from
the reference alignment defined by the Main Street forming clusters of local grid are similarly found in all the sectors.
The grids are not only continuous within each of the two sections of the HR Area and the VS
area that are separated by about a distance of 90 meter, they are also part of the same grid system, that is the
two
grids coincide when extended as they belong to the same reference grid that aligns with the Main Street. It is also found
from
the
measurement of the aerial photograph that the two areas follow one reference axis along the east-west direction. This continuity of the grid towards the
northern
end could not be examined
due to the unavailability of the drawing of that part of the city.
The Pattern of Plots and The Coaxial Field Division
Among the transverse
(east-west) and the longitudinal (north-south) boundaries of the cluster blocks, the later is comparatively more
pronounced
in its continuity. The plot of the blocks are divided
along the lanes that run parallel to the main street. The longitudinal divisions, in general, define the depth of the cluster blocks which is most evident in HR and VS Area.
The transverse divisions that give the width of the blocks show a certain degree of variation. In the figures the width
of the blocks whose boundaries' closely coincide with the superimposed
grid are not separately noted.
Measurement values are noted for those blocks whose boundaries donot coincide with the grid. It can be seen that most of the cluster
blocks have width from
9.6 m to 19.20m. while
a depth of 38.40 m are not uncommon in all the three areas. The dimensions of the
blocks not corresponding to the standard
grid width is usually due to the width of the lane, which, in all the instances, is laid
at one side of the
grid line.
The reference grid, which
is
aligned
with the eastern
boundary of the Main
Street has revealed a vital feature on the manner of the layout of the streets, lanes and the plots. The alignment of the lane and plot boundaries in this reference
grid shows that the survey measurement was not done by measuring the segments from
centre
of one street or the lane to the other. The greater continuity of boundaries
in the longitudinal directions suggest that these divisions
were primary in the settlement layout. It has been found from the measurement of the aerial photograph of the site that the distance between the reference axis of the Main
Street and the
corresponding side
of the
street or lane of the citadel area at the west is close to the multiple of 19.20 m (416 m = 22 x 18.9 m) (Fig. 2). Likewise
the
corresponding distance
to the Main Street of the DK area is also close to the multiple of the same grid (193 m = 10 x 19.3 m). This in conjunction with the layout manner of the streets and lanes and the close alignment of the boundaries
of the blocks demonstrate that the area was first regularly divided into longitudinal sectors of width of
19.20 m
running north-south. We are not certain
whether
this division long preceded the beginning of the settlement or was necessited as a first step for the foundation of the settlement. It has been shown in the
case of British isles that extensive tracts of land were divided in prehistoric times
going ba k upto third millenium in a manner that has been referred as'coaxial field system'where land covering more
than one hundred hectares are found to have been divided on equal intervals with continuous
longitudinal boundaries that follow one reference axis.11 The case of Mohenjodaro
shows a close affinity with this system
of field division. The layout of streets and lanes of the city complies with the existing field boundaries instead of their
obliteration. This is particularly clear from the instance that the blocks
at the east of the Main Street in HR Area is not of usual depth formed by 19.2m wide grid due to the extra width of the main street which is
around half a grid
wide. Thus the
depth of the plots were defined by the existing net of field boundaries and the lanes drawn within the grid, and not in a manner where blocks are laid in uniform
size.
The Modular Dimesion
It can be seen from the three areas of Mohenjodarn that the dimension, which frequently occurs in major
cluster blocks is 19.20 m. Likewise the most common dimension
of the minor individual blocks is 9.60
m which is a half of 19.20 m. It is to be noted that 9.60 m is also the width of the Main Street. In addtition, there are also blocks whose widths lie in between the above two values. The coincidence of lanes with the grid of 19.20 m shows that the dimension of 19.20
m is determinant in the layout of settlement blocks and street planning. Various other dimensions found in the width of the cluster blocks show that the values are a
common multiple of a standard unit measure which is equal to one danda. This danda, however, is a measure that we know only from Kautilya of 4th century BC. He gives the value of one danda as 4 hasta where one hasta is of 27 digits. Although there have.been various estimation of the length of one hasta, there are reasons
that its value �pproximately is 48.5 crn12. This measure gives approximately 1.8 cm
to one digit and agrees more closely than other neighbourhood values with the size of the
bricks of the early settlement period, which is given as 10 by 5 by 2.25 inches. With this measure
of a digit and hasta, a danda will be of 194 cm length. Most of the dimensions of the
cluster blocks of the settlement as well
as the
grid can be derived as the multiple of this danda (Fig. 3). In particular, what is to be
emphasized is that 9.60
m and 19.20 m are 5
times and 10 times of one danda (5d and 10d). In Kautilya, ten times
the length
of a danda is defined as one rajju. Rajju in Sanskrit means chord or rope, and is also mentioned in Jatakas where the land surveyors are mentioned as rajjugahaka. Leaving further details on the records of the early historic documents here, what we can be certain with·the·available data is that such system of measurement and survey existed and was employed by the peoples of Mohenjodaro in their town planning. The dimensions fit perfectly
to be otherwise. Accordingly, we find that most of the lanes of the
settlement are around a width of 1 or 2 danda.
The
ca況of Sirkap (Taxila) and Thimi Sirkap
While Mohenjodaro is known as one of the two great cities of prehistoric Indus Valley civilization belonging to Chalcolithic period, Taxila, now known as Sirkap, is said to have been built sometimes during the second century BC.13 Sirkap is one of the three separate settlement areas discovered within the Taxila valley that is irrigated by several tributaries of River Ravi (Fig. 6). Bhir Mound, a settlement at the southwest across the Tamra River, a tributary of Ravi, is thought to be older to Sirkap, while another settlement, Sirsukh, at the northeast across the Tundi Tiver was a town built later in the second century AD ascribed to the Kushana king Vima KhadphisesりThe excavated area of Sirkap reveals a settlement enclosed by ramparts within which there is a town at the lower plain and an elevated hill terrace which has few monastery complexes. This area is also thought to be a place for the citadel. An inner rampart in between the upper hill terrace and the lower settlement has been revealed through excavations of 1945.
The
town of Sirkap reveals a planned settlement laid out in a regular pattern of streets (Fig. 7).
It has one main street that runs north south which is almost at the centre of the settlement area so far revealed by the excavation. The town is then divided into blocks of equal width by lanes that run east-west. The lanes are through and cross the main street at right angles. The regularity in the size of the quarter blocks breaks at the central sector of the settlement and at the north from where a structure is designated as'Palace': The dimensions of the division of the quarter blocks from the north to the south is given in Fig 1. The average width of the blocks at the northern section where there is a consistent pattern of division is close to 38.5m, which must be the standard planning module for the division of the settlement blocks. The quarter blocks at the north with regular pattern of division are then analysed to study the division of the blocks into plots and thereby derive standard measures employed in the planning of Sirkap. The quarters are Al to Cl at the east and A2 to H2 at the west of the main street. The analysis follows the same procedure as employed to Mohenjodaro.
The
continuity of the boundaries of the building clusters in these blocks shows the longitudinal division of the quarter into two halves and then this northen and southern halves
have further divisions, which in most probabilty indicate the residential plots. In the main, there are two types of cluster blocks--of 19.2 m and of 9.6 m width. There are also blocks of width between 9.6 m and
19.2 m, but such cluster blocks are comparatively fewer than those of 9.6m. Likewise, blocks larger than 19.5 mare also not_apparent in the formation of cluster blocks. The
occurance of smaller blocks of 9.6 m is also
rare. Moreover, the
boundaries of the cluster blocks giving width
of 9.6 m and
19.2 m cross the longitudinal halves and even the quarter blocks as shown by the dotted lines in the figure suggesting the above two dimensions as the frequently utilised measure
in the divisions
of the plots. The three measures
of 9.6m, 19.2m and
38.Sm that are most frequent in making the
divisions of quarter blocks and in the size of cluster blocks within the quarter blocks in the town shows that 9.6 m is the basic standard unit for the division of the
dwelling plots. Interestingly the measure of 9.6 m, as in Mohenjodaro,
is also the width of the main street of Sirkap. A lower multiple of this unitis not readily apparent in this analysis of the cluster blocks.
But the dimensions of quarter blocks that differ from the standard width of 38.5 m, which are at the north of the temple precinct, gives us a modular unit of 1.94 m—the measure of a danda that was the standard
measuring module
of Mohenjodaro.
The widths of the lane between blocks, which are of regular size, range from 3.5m to 4.2 m suggesting that 2 danda was the planned
width of the lanes. Thus, if one takes the pattern of cluster blocks in the most regular quarter blocks illustrated by the northernmost blocks, a basic standard plot in a quarter block after deducing the width of the lane comes to be 9
x 5 danda.
The
study of the aerial map (Fig. 8) of the.neighbourhood fields around the walled
area of the city both inside and outside shows a division pattern of the fields in the same manner as is found in the layout of the lanes of the town settlement, i. e.,
the land is parcelled into plots of 19.5 m width where, in many instances, further subdivisions are to be found. The divisions are to be found not only in the directions of the lane but also longitudinally that run parallel to the Main Street. The regularity of the field division at the western field within the wall is there all throughout from north to south despite that the lanes
at the southern part of the town are found at intervals giving different block widths. Thus the pattern of the surrounding field division shows that the
principle in the division of the settlement blocks and the fields was one and the same. What we are not certain is whether the land division long preceded
the town layout or followed it along the same line. Given that Taxila was an inhabited area many centuries prior to the described period of building of Sirkap, and was the capital centre of the most rich Satrapy of Achemedian empire, it is certain that the valley basin was intensely cultivated supporting the need of the capital and its role as one of the important cultural and commercial centre of the region. As
has been described earilier, the measuring system used in Sirkap and the surrounding fields was already known during the time of Maurya rule long before the Bactreans arrived in Taxila.
Yet another even more important bit of evidence to the earlier
existence of this system
of planning in Taxila comes from the settlement of the Bhir
Mound itself, which, according to the findings of archaeological artifacts, has been described as the earliest settlement of Taxila. The settlement has been described as irregular and haphazard
by Marshall. His opinion with regard
to the form of the settlement rests upon the limited area of excavation and the pattern of streets revealed from the upper stratum. But fortunately, wall lines of of the building clusters in the lower stratum are also shown in the drawings (Fig 9). One can observe
from
this drawing that the earilier structures shown by dotted lines
have more regularity
in their layout; but which
considerably changed by the building of the later phase superimposed upon the structures of the early period. This is particularly apparent in the blocks between the First and Third Street. The superimposed grid
shows a number of blocks in
continuity with
a width of 19.2 m. The blocks that follow the grid
cover the greater part of the excavated area lying between Second and Fourth Street, in addition to the two blocks at the north of Square S in the Third Street.
Further, the width between the First Street and the Second Street is also 19.2 m. The division of the 19.2 m width blocks into half are apparent in the blocks between Fourth Street and First Street. We consider that the findings of such regular size of blocks within this small area is significant to establish a definite relationship between the settlement in Bhir Mound and the pattern of field divisions
in
Sirkap neighbourhoods as well as the planning of the settlement there. It demonstrates that the land
in Sirkap was divided for its cultivation prior to the building of the settlement, and if the planned settlement began only by second century BC, it followed the pattern
of land division that already existed long before. Further, it may be worthwhile to note that the block size of Bhir Mound is same as it is in Mohenjodaro while that of Sirkapis twice this width.
Thimi
The pattern of street planning similar to that of Sirkap is also found in one
of the ancient towns of Kathmandu Valley which is Thimi (Fig. I 0). In fact, the grid of 38.5 m
was first discovered in
Thimi which lead to the search of such planning modules employed
in other
areas as well. The layout of Thimi follows the topographic features of the land which is situated in a plateau extending from north to south and then descends to one of the tributary stream of Bagmati River of the Valley. The Main Street of the settlement runs north-south and the settlement is divided
into quarter
blocks by parallel lanes running east-west that branch from the main street. The analysis
of the building clusters, remnant of lanes and the measurement survey of the building frontages along the length of the Main Street from the southern end to the northern end of the early settlement core showed that the lanes were laid at an interval of 38.5
m. The town developed through a number of phases, and historical documents record its existence prior to 1350, but Licchavi period (2nd C. AD C. AD) inscriptions found in the settlement and other features suggest that the town has been there long before this date. One of the earliest structures of the settlement is the'palace'and lies at the central
part of the eastern sector of the planned area. The regularity in the planned form, through the longperiod of continued settlement, is considerably effaced but is still possible to see its early planned features. The finding of this module in the town settlement was aided by an earlier discovery of an area that lies at the lower basin right at the east of the town plateau. This area which is now a vegetable garden of the residents of the town is regularly divided into 9 sectors whose average width is 38.5 m. However, there are no remains that can be observed at the surface
within the area. The coincidence of the division pattern in the lower basin and the town suggested a possibility in the prevalent use of this module in the survey and planning of settlement in the region.
The unavailability of the measured drawings of the dwellings of the
settlement clusters, do not, at present, permit us to analyse
in precision the plot division patterns within the quarter
block of Thimi. The
study of the clan settlement clusters
however shows that the quarter block was divided into homestead
type plots with their main entry access from
one of the lanes. The division pattern
of the lower basin shows that there is first the longitudinal division of the sectors
which
are then
further divided
into a number of plots. This pattern of division follows
modular dimensions
and shows similarity with the
general
division pattern
of Sirkap.
From Mohenjodaro to Sirkap and Thimi
The above analysis of the settlement of Thimi, Sirkap and Mohenjodaro shows obvious links between the three towns in the employment of a standard measurement rod. What is strikingto note here is that it is not only the danda unit, but more larger units, which are of 10 danda and 20 danda magnitudes, are
common to each of these settlements. In addition, of equal importance to note here is that a hasta unit, which has been utilised in wide geographic and cultural regions, is a measure that could vary
considerably unless a
standard measure is handed down in continuity as an outstanding tradition. For instance, in ancient Egypt, this measure is given two different values of 46 cm and 53.5 cm, known as the common cubit and the royal cubit.15 Likewise, Kautilya gives three different measures of cubit which range from 24 digit to 27 digit that give values of hastas somewhere ranging from 43 cm to 48 cm. The length of each hasta type varies again according to the width of a digit, which is said to be the width of the middle finger. Such disparity in the hasta measure gives up to 2 m of difference in a length of 20 danda. But the multiple dimensions found in these settlements are close enough to prove that there
is a continuity in the surv.ey measurement tradition from Mohenjodaro to Sirkap and Thimi. It is surprising to note that not only the standard hasta measure of Indus comes to Sirkap but its standard multiple units which are danda and rajju are also inherited. The size of the standard plot also shows similar continuity. It is therefore only natural that the main street both in Mohenjodaro and Sirkap have the same width of 10 danda.
If we leave this fundamental convergence that exists among the three settlements, the differences in the manner of the division of the plot between the two settlements of Sirkap and Mohenjodaro are also apparent. In Mohenjodaro, the excavated areas have not revealed
equal size of quarter blocks in between the lanes although it is
to be noted that the more important lanes are in variably at a grid of 19.5 m. The lanes are often not through, connecting from one main street to the other, but dog-legged. Short cul de sac lanes
are often found leading to the interior of the blocks from the Main street. The lanes are more narrow, almost half the width of the lanes of Sirkap. In contrast, the lanes of Sirkap are generally through and there are no access lanes
to enter the plots of the quarter block from the main street.
The town of Thimi in contrast to Mohenjodaro and Sirkap, which are only the remains of historic and
prehistoric past, is a living settlement. But it has affinities both with Sirkap and Mohenjodaro. While the grid
of 38.5m is common to Sirkap, the Janes of Thimi including other
towns
of Kathmandu Valley, when not through, show a pattern similar to that of Mohenjodaro or that of Bhir Mound, which
turn
at right angles than
going diagonal. The diagonal links are common in later Islamic settlements
and prehistoric settlement excavated in Ur and and in Mediterranean
region
such as Mycenian towns.
As the existence of the danda measure is found in both Mohenjodaro and Sirkap, it is of significance to find that this modular measure also existed in Kathmandu Valley. This evidence is unequivocally
furnished by one ancient Buddhist monastery known as Chabahil (Sanskrit—Charumati Vihara) (Fig. 11). This Buddhist monastery according to Nepalese chronicles, is one of the oldest extant Buddhist monasteries in Kathmandu Valley. Its plan is typical of those early monasteries that can be found in the remains of India, or Gandharan region of Pakistan including those built in Sirkap, where monastic cells are arranged around an open quadrangular courtyard. The figure shown is the measured drawing drawn by Nippon Institute of Technology.16 lt can be
seen that the timber columns of the peristyle courtyard have the span of 1.95 cm measured centre to centre. This dimension is equal to four hasta or one danda
with a hasta measure of 48.6 cm. The multiple of this dimension gives other values such as width of the open court and also the internal clear width of the rooms. The overall extenal dimension of the courtyard building is also the multiple of the
danda,
but it is due to the width of the brick masonry wall which is roughly one hasta wide. Not all monasteries built later confirm to this standard module. Therefore, it is in a way fortuitous to have before us the measured drawings of this ancient foundation that confirms to the above danda measure.
And this evidence alone sufficiently demonstrates that the standard module that is derived from one particular hasta system of measurement, was employed in Nepal from ancient times. It is
exactly this measure which perfectly fits to the multiples of 10 danda, 20 danda and other intermediate multiples that are found in the street patterns of Thimi as well as those of Sirkap and Mohenjodaro where the plot divisions also follow the same modular units.
John Marshall who worked in the excavation of Sirkap for twenty years from 1913 to 1934 assumed that Sirkap was planned by the Indo-Greeks when they ruled from the last quarter of second century BC for about one century (189-90 BC). Although Mohenjodaro clearly exhibited a system of planning both in its form and other civic infrastructures as
is readily seen from the regular orthogonal patterns of streets and layout of drainage and sanitary disposals, he nevertheles believed that only the Greeks did have the tradition of planning the city in a chessboard pattern. However, as there were no evidence to unequivocally support this argument, his own reports written at different times contradict with each other.
Conclusion
The
present study for
the first time demonstrates that Mohenjodaro was a city based on certain definite planning principles to guide the division of settlement quarters and dwelling plots. The layout was done upon a'coaxial system of field division'whose principles of measurement were followed down to the layout of streets, lanes and the individual plots. The planning grid encompasses the length and breadth of the area defined by the excavated sections. It emplyed a measurement system which has been described by Kautilya as late as 41h centurry. What is surprising is that the
planning module employed in the Indus city of Mohenjodaro, Sirkap of Gandhara and Thimi of Kathmandu Valley are the same. The
coincidence cannot be by chance as has been shown by the exactness of their confirmity to the standard multiples derived from one particular system of hasta measure. One of the link between these widely separated settlements is to be seen in Kautilya himself who lived in Taxila and Pataliputra, the capital of Maurya empire, and who
wrote exactly on this system of measurement in his economic treatise Arthasastra. Nepal was already in communication with Maurya empire of those times as is known from the same treatise.18
However Kautilya is not the ante terminus, but only the post terminus on the time of the communication link between Nepal and Indus civilization as well as Sirkap. What is seen from the analysis of the settlement plans of these areas
is that the Mohenjodaro tradition of the [ndus people on town planning and
survey was inherited down to later historic periods and spread widely in the region, where
Kathmandu Valley still preserves those elements in architecture and settlement plans.19
The
next link of concern is between the early historic settlement of Sirkap and the Chalcolithic city of Mohenjodaro. It is to be bear in mind that the archaeological reports on the remains of Taxila including the Bhir Mound have not yet discovered the material that can definitely date the foundation of these settlements. The dates only show the period of their existence but not the time of their first settlement.
However, later archaeological works in the region of Taxila, for instance, that of Saraikhola20 one and
half miles
from
Sirkap, and late findings in the Hathial Mound a few hundred meters away from Sirkap, have now given the evidences of the settlement in Taxila contemporary to Indus period. Allchin describes the finding of pottery sherds in the Hathial
Mound area that belonged to early Indus period as well as those of later periods but earlier
by several
centuries to those of Bhir
Mound.21 Thus recent archaelogical reports have begun to establish a continuity of settlement in Taxila from as far back as
Indus period. Our own analysis of the structural remains of Sirkap shows its link with the Bhir Mound and Mohenjodaro tradition of town planning. Thus if the continuity in the inheritance of the town planning idea from Indus to Sirkap and Thimi in Kathmandu Valley is evident as this paper shows, it also appears from other findings that there was also the continuity of the people themselves
who built their
settlement upon
this tradition
in
Taxila.
As to the relationship upon
the
origin
of the
people of Kathmandu with those of Indus, it may not be a surprise if the future
archaeological studies
establish a positive link between the two people.
References
Allchin, F. R., 1982. How old is the city ofTaxila? Antiquity, vol. LVI: 8-14.
Fleming, A., 1987. Coaxial Field Systems: Some Quenstions of Time and Space. Antiquity, vol. 61: 188-202.
Funo, S. et al., 1997a. "Street Pattern and Block System of Cakranegara City, Lombok, Indonesia: Spatial Structure of Cakranegara,
Part!", J. Archit. and Plann. Env., All, 491, Jan, pp. 135-139.
---- 1997b. "Street Pattern and Block System of Jaipur City, Rajasthan, lndia--An Analysis on City Maps (1925-28) made by Survey of India. Part I."
J. Archit. and Plann. Env., AIJ, 499, Sep, pp. 113-199.
Ghosh, A, 1948. Taxila (Sirkap), 1944-45. Ancient India, No. 4.
Halim, M.A., 1972a. Excavations at Sarai Khola, Pt.1, Pakistan Archaeology, 7, pp 23-89.
---- 1972b. Excavations atSarai Khola, Pt. 2, Pakistan Archaeology, 8, pp 1-112.
Havell, H. G
Kangle, R. P., (tr.). The Kautilya Arthasastra, Motilala Banarasidas, Delhi.
Lal, B. B. and Gupta, S. B., ed., 1984. Frontiers of Indus Civilization. Indian Archaeological Society, New Delhi. Marshall, John, 1931. Mobenjodaro and Indus Civilization, Vol I-III. Arthur Probsthein, London.
Marshall, John, ed., 1951.Taxila, Vol I-III. University Press, Cambridge.
Marshall, John, ed., 1960. A Guide To TaxiJa.
Mainakar, V. B., 1984. Metrology in
Indus Civilization in'Frontiers of Indus Civilization',
B. B. Lal and S. B. Gupta, eds.,
Indian Archaeological Society, New Delhi.
Nakagawa, Tak-eshi, 2000. Planning Method of Angkor Wat.Third International Symposium on Architectural Interchange
in Asia, Cheju National University, Cheju, Korea, Feb, 23-25, pp. 175-185.
Nippon Institute of Technology, Research Mission,1986. The Royal Buildings and Religious Remains of Nepal.
Pant, Mohan,
2002. Spatial Formation of Kathmandu Valley
Towns—The Case of
Thimi. PhD dissertation submitted to Kyoto University.
Peterson,
J.M., 1990. Why did the idea of co-axial field system lastso long? Antiquity, vol. 64: 584-591. 1947. Sishupalg.arl1. Ancient lndia. Vol. 5
Punini. India in Panini―5、「imes
Rao, 1980. Lothal, Vol I-III. Memoirs of Archaeological Survey of India.
Tiwari, Sudarshan, 2000. Ancient Architecture of
Nepal. Vaastu, vol. 2, pp 67-72. 1978.度 の 史
Vij, B.B., 1984. Linear Standard in the Indus Civilization in'Frontiers of Indus Civilization', B. B. Lal and S. B. Gupta, eds., Indian Archaeological
Society, New Delhi.
 |
1 It was the finding of seals with legends inscribed in Indus script that led Banerjee, when he was digging the Buddhist remains in 1922, to conclude that the site belonged to Chalcolithic age. The discovery follows that ofHarappa in 1921 by
D.R. Sahni. See Marshal,
J., 1936, ppl0-11, Lal, B. B. and Gupta, p x)
2 The work published in three separate volumes is contributed by aurthors who led the excavation works at different areas of the city. The first volume gives the general introduction of the site, description of each areas, architecture, religion, figurines and potteries. The second volume are on coins, weights and measures, and Indus script. The drawings of individual buildings, excavated sites, and photographs are compiled in the third volume.
3 McKeys, one of the excavators of Mohenjodaro writes on the town planning of Mohenjodaro:
That the people of the Indus Valley civilization had some idea of town-planningis shown by the regularity with which the city of Mohenjo-daro is divided up, a regularity which is striking for an ancient city of the east, or for that matter, would be in a western city to-day. We find streets,
both wide and narrow, on the whole successfully aligned, instead of windingin the way that was usual in most early cities. Indeed, from the very systematic way in which
the
city is laid out we must conclude that it was methodically planned and not just built haphazard.
All the main streets, so far excavated, are oriented to the points of compass, and what little deviation there is is so slight that this orientation is very conspicuous. Houses and public buildings correspond in their orientation with the streets.... One of the two procedures must clearly have been
followed: either the city must have been
built on a defmite scheme from the commencement of its history, which followed perhaps on a compulsory move from another site; or the arrangement of the city was the deliberate reconstruction on town planning lines by order of the city fathers or a higher authority….In town-planning the peoples of the Indus Valley appear to have been superior to the early Sumerians............................................................................................................... '
Architecture and Masonry, pp 282-83 (Marshall, J. ed., Vol. 1, 1936)
4 Marshall, J. ed., 1936.
5 The
stratums of Sikap city are classified into 6
levels—I to VI from top to bottom.
Marshall assigns the lowest two levels to the Bactrean period, IV
and III to Saka, and II and I to Saka Partheans and Kushana respectively.
See Ref. 10.
7 Kalibangan is located at about 1600 km southeast of Harappa. The excavation has revealed
the settlement made up of two fortified areas. The upper area is thought to be the ritual centre while the lower one the main settlement area. The town is divided into sectors and blocks by north-south streets that are parallel
and laid at regular intervals, while the east-west streets are not always through as the north-south streets. The layoutis a quasi grid in the term ofT. Ohji. See
Lal, B. B.,1984.
8 See Ref. and Ref. Mainakar studies the relationship of the scale,
carved in an strip of ivory, found in Mohenjodaro, and that of Lothal, which was carved in shell, with respect to the size of bricks, the plan of Great Bath of Mohenjodaro and the Dock of Lothal. While Vij makes a comparative analysis of weights and the dimensions of the Great Bath with respect to its implication of the knowledge of pi and astronomical instruments.
9 Marshall first published'A guide to Taxila'in 1918 which was subsequently republished in 1936 and revised in 1960. It
was probably conviction
on the
hypothesis of catastrophic theory of demise of Indus civilization, or the wholesale destruction of the Indus people by the invading Aryans (Wheeler) or other nomads from the
north (Marshall)
that any search on the relationship of Sirkap to Indus culture was foreclosed.
However, it is difficult to understand Marshall that when he with all the documents to speak on the planning features of Mohenjodaro, and Sirkap by 1960, still prefers to
remain silent on Indus when writing on the town of Taxila.
lO The excavation in Sarai Khola, 2.5 km from Taxila, showed early Harappan or Kot diji type potteries (Halim, 1972).
On the finding of Indus Cultural artifacts near Sirkap Allchin, 1982.
12
l-ITaxila was the capital of Achemedian empire from 61h century to the later quarter of 4'h century BC when they were
overrun by the army of Alexander in 325 BC. Alexander soon returned
to the western territories
and his viceroys ruled only for 3 years in Taxila before they were driven away by the forces of Chandragupta in 323 BC. The Mauryan empire from Pataliputra ruled Taxila for around one and a quarter century whence it was the seat of viceroy prince Ashoka and later his son Kunala. In 189 BC the
Bactrean came to rule in the Gandhara
region
with Taxila as their capital. Following them, Sakas (90 BC-78 AD) and Kushanas (78 AD-320 AD) ruled Gandhara untill the advent of Gupta empire. The archaeological remains and artifacts as well as the literary records all show that Taxila was one important centre
of Buddhism. Panini (5th century BC), the celebrated grammarian
of Sanskrit, and Kautilya the
teacher
of Chandragupta, lived and taught in Taxila.
15 See measurement history.
16
17 Since many authors following him continued to echo Marshall's view, we consider it relevant to quote some of the relevantparts at length:
、、The Sirkap city, built by Bactrian Greeks in the opening years of the second century BC., occupies
the extreme western
spurs of the hathial ridge with the
small well-defined plateau
on their northern side, and
included
in its northern suburbs the area called Kaccha Kot—from the fact that it was defended
by earthen ramparts only, a section of which may still be seen enclosed within a bend of the Tamra nala. Excluding this suburb, the city's walls were nearly 3.5 miles long -a third again as long as those of the older city, and were built of solid, coursed stone rubble instead
of mud brick. In
accordance with the Hellenistic principles of defence, they included within their perimeter a considerable area of hilly ground as well as an isolated acropolis, and in other respects the city was laid outon the typically Greek chess-board pattern, with streets cutting on another at right-angles and regularly aligned
blocks of buildings. Notwithstanding tha the city was several times destroyed and rebuilt and that many transformations were made in individual
buildings, this Greek layout was on the whole well preserved
down to the latest days of the city's occupation." (Marshall, 1960)
Elsewhere, he again writes on Sirkap: "…From every point of view it was just such a site as might be expected to appeal to a Hellenistic town-planner, who required, first and foremost, a considerable space of level ground, if he was to design an effective layout of streets and buildings on the schematic chessboard pattern which was then in fashion; in the second place, a backing of low defensive hills, some part of which could be brought within the city's perimeter; in the third place, the presence of one or more streams alongside the walls to ensure a steady supply of water and at the same time increase
the strength of the defences; and,
in the fourth place, an isolated hill suitable for an acropolis."
However,
Montiiner Wheeler, following the new report of A. Ghosh on the excavation made in 1945 makes the following
reference to Marshall's 1936 report, which according to him, needs modification:
At one or two points in the northern part of the walled city, Marshall dug down in small areas to the natural soil. He found 'remains belonging to six successive
periods of habitation.....represented by clearly defined foundations of rubble masonry, with layers of debris above and below them. Of these successive strata of buildings, the
fifth and sixth from the
top belong to the period of Greek rule at Taxila (c. 190-85 BC) The fourth city belongs to the time of the early Saka
kings, probably of Azes I, many of whose coins were found buried in small hoards beneath the house floors.
It was this same Saka King who was responsible for contracting the city's perimeter and substituting well built walls and bastions of solid stone in place of the older fortifications of mud, and was responsible also for the symmetrical lay-out of streets and lanes which continued to distinguish it to the end of its history. The third city from the top, which is less clearly defined than the others is also referable to the period of Sakas.
The second city dates from parthian from Parthian times in the first half of the first century AD., and is characterised by the use of diaper masonry along with the ordinary local rubble. It is to this city that most of the structures exposed in Sirkap belong.'
What the excavations of B. Ghosh suggested was that, the city walls belonged to later phase (ascribed to Saka) than those of the period indicated by 6th and 5th strata (ascribed to Indo-Greeks), and that the high ground, which was suggested as acropolis, was
only incorporated later when the stone wall was built around the city.
Marshall's own contradictory reports on the authorship of Sirkap and Wheeler's support to the works of B Ghosh shows
the unfounded assumptions on the lndo-Greek planning of Sirkap.
18 The mention of Nepala comes in the chapter、The Activity of The Head of The Departments': The black
Bhingisi,
made of a collection of eight woven strands (and)
the apasaraka
which keeps
off rain,--that is (woolen cloth) from Nepala. (2.11.100). Kangle, R. P.,
1992.
19 One more instance of a planned city of early historic period is Sishupalgarh in Odissa. The settlement area is enclosed
by earth ramparts with two gates in each direction. The measurement done by the author on the aerial photograph shows that the distance between the two gates of each direction is exactly the same and is eleven times 38.5 m. The intersecting lines joining the centre of the oppositegates make exactly 90 degree angle between them showing a great degree of precision on the survey work.
0 件のコメント:
コメントを投稿