Content uploaded by Marloes H.N. Bakker
Author content
All content in this area was uploaded by Marloes H.N. Bakker on Mar 12, 2019
Content may be subject to copyright.
SPECIAL ARTICLE
Economi c & Politic al Weekly EPW january 9, 2010 vol xlv no 2 45
Ra shmi Kiran Shre stha (rashmi777@hotmail.com) and Rhodante Ahlers
are with UNESCO-IHE, Delf t, The Netherlands. Marloes Bak ker is with
the Cooperative for Climate and Water, The Hague and Joyeeta Gupta is
with UNESCO-IHE , Del ft a nd Univ er sity of Am ster dam,
The Netherlands.
Institutional Dysfunction and Challenges in
Flood Control: A Case Study of the Kosi Flood 2008
Rashmi Kiran Shrestha, Rhodante Ahlers, Marloes Bakker, Joyeeta Gupta
The Kosi f lood disa ster of 2008 in Bihar and also in Nepal
high lights two key issues re lating t o flo od cont rol. Th e
first is the failure of the structural approach to flood
control on the Kosi and the second is institutional
dysfunction with respect to trans-boundary flood
management. This article discusses the key reasons for
the failure of flood management in the Kosi, through
stakeholder interviews and observations in the
aftermath of the flood. The institutional context
comprises several challenges such as trans-boundary
politics between Nepal and India, the internal politics of
Nepal, intra-state politics in India, the inherent
weaknesses of the Kosi treaty, structural flood control
strategy and the lack of connection between
governmental decision-making bodies, implementation
agencies and civil society.
The Kosi floods of 2008, caused by an embankment breach
at Kusaha village of Sunsari district in Nepal, was disas-
trous. The embankments were built by India in 1959 as re-
quired by the 1954 Kosi treaty between Nepal and India. The
treaty provided for construction of embankments in Nepalese
territory to control flooding both in Bihar state within India, and
a section of Nepal bordering with India. The recent floods raise
two main issues: (1) Are the flood control measures appropriate
for t he u nique nat ure of the Kos i ri ver? (2) To wh at exte nt can t he
flood be attributed to the institutions managing the Kosi river?
This is of critical importance if similar floods are to be prevented
and/or man aged bet ter in t he future.
The unique characteristics of the Kosi river and existing flood
co nt ro l m ea su re s have be en ex te ns iv el y d is cu ss ed by D ix it ( 2009 )
and Sinha (2008) and also by Kale (2008), Reddy et al (2008)
and Gyawali (2008).1 However, although all authors refer to
the role of the institutions involved in the management of the
Kosi river, not one analyses them. The research presented here
seeks to a ddress this gap by assessing the role of the institutional
m echanisms established to deal with the challenges of the Kosi
river under the Kosi treaty. It focuses, in particular, on the
Nepali perspective.
This article is based on field research carried out between
November 2008 and January 2009 in Nepal. Three Nepalese
villages (Shreepur, Haripur and Paschim Kusaha) severely
affected by the 2008 Kosi embankment breach were visited. In
addition, 50 semi-structured interviews were conducted with
different stakeholders including government officials and flood
victims in Nepal. Four interviews were held with Indian officers
from the Kosi project office in Birpur, Bihar, the liaison officer
from the water resource department from Bihar, and a senior
official from the embassy of India in Nepal.
This paper first reviews the impact of the Kosi embankment
breach of 2008 along with the key points made in the literature.
This is followed by an analysis of the relevant institutions
managing the Kosi river. Finally, it concludes with suggestions
for better flood management in the Kosi basin.
The Kosi F lood of 20 08
On 18 August 2008, the Kosi river breached its embankment and
displaced 45,000 people from three severely affected villages
(Haripur, Shreepur and Paschim Kusaha) of the Sunsari district
of Nepal (minutes of the United Nations Office for the Coordina-
tion of Humanitarian Affairs-UN OCHA, 29 March 2009).2 About
3.065 million residents from 1,704 villages in north Bihar were
similarly affected (Mishra 2008b), and around 4,648 ha of
SPECIAL ARTICLE
january 9, 2010 vol xlv no 2 EPW Economi c & Politic al Weekly
46
in different locations (see Mishra 2008 for details). It could be
argued that instead of controlling flooding, the embankments
have aggravated the conditions in the Kosi river (Mishra 2008a
and 2008b; Sinha 2008; Gyawali 2008; Kale 2008; Dixit 2009).
The embankments gave a false sense of security (Mishra 2008a;
Kale 2008; Sinha 2008; Gyawali 2008; Dixit 2009).
Nevertheless, the 2008 flood is different from previous floods
in that past embankment breaches were mostly accompanied by
high discharges along the Kosi river. In 2008, the embankment
breached during a period of below average discharge compared
with the same time period in previous years (I4, 2009). 4 There-
fore, this is not a flood caused by high discharge brought with
Asian monsoon, cloudburst and landmass movement from the
Himalayas. In addition, the 2008 flood breached in Nepal and for
th e fir st time up str eam of the bar rage. Severa l aut hor s argue that
a lack of priority and urgency in maintaining these structures
contributed to the embank ment breach in 2008 (Mishra 2008a;
Th akka r 2009: see end note 2; Di xit 2009).
Hence, Dixit (2009) calls for a paradigm shift from structural
flood control to flood resilience. This is echoed by Sinha (2008)
who calls for a move to flood management instead of flood control
while Kale (2008) calls for better regional cooperation for im-
proved flood management in the Kosi river. Another issue these
authors signal is the management of the trans-boundary Kosi
project for flood control. Dixit (2009) argues that institutional
dysfunction and governance deficit allowed the 2008 flood to
h appen. Gyawali (2008: 5) refers to the Kosi flood as the outcome
of “an unholy marriage of wrong technological choice, bad institu-
tional arrangements and half a century of political mis conduct”.
Moreover, Himanshu Thakkar (2009)5 and Mishra (2008b and
2008c) argue that the Kosi embankment breach is the outcome of
negligence of duty and lack of accountability at different levels.
In sum, the literature on the Kosi highlights two challenges –
the technological choice and the institutional design. This paper
elaborates only on the institutional dynamics of flood control
along the Kosi river. First, we show the lack of awareness, warn-
ing and preparedness to deal with flood risks along the Kosi river.
We then present the complex institutional and communication
mechanisms of the Kosi project and highlight the institutional
challenges in managing the Kosi river. Finally, this paper identi-
fies how lessons learned may be used to improve the manage-
ment o f t he Kosi river in the future.
Reasons for the Severity
Lack of Awareness, Warning and Preparedness for Flood
Disaster: The fear and rumour of embankment breach starts with
every monsoon in the Sunsari and Saptari districts of Nepal where
the embankments begin. The flood victims in Nepali villages ex-
plained that: “there have been rumours of probable embankment
breach for many years but it did not breach so we did not believe it
until it happened” (Shreepur village, 18 December 2008). A flood
victim living near the breach pointed out: “in the evening prior to
the e mbankment breach, villagers told me about rapidly weath-
ered embankments; when I went to see the location, the river had
touched the eastern embankments and the water level was rising,
a gricultural land and crops were washed away in Nepal (minutes
of UN OCHA, 29 March 2009). Sand deposited by the flood ranged
from three to seven feet (personal observation) and will hamper
agricultural activities for many years to come. At the embank-
ment breach in the Nepali village of Kusaha, the landscape ap-
pears desert-like. Visits to the shelter camps in Nepal provided
graphic images of the devastating impact of the flood on its vic-
tims’ emotional, physical and economic well-being.
The damage caused by the Kosi flood of 2008 is the highest in
five decades of flood history in Bihar (Kale 2008) and according
to a Nepali senior officer from the Ministry of Home Affairs and
member of the Central Disaster Relief Committee, the worst in
the entire flood history of Nepal (I3, 2009). 3 The event rai se s sev-
eral questions: Why did the embankment breach in 2008 lead to
such a devastating flood? Is this likely to be a one-off event, or an
indicator of what is l ikely to happen more of ten in t he future?
The Kosi r ive r is des cri bed a s a u nique riv er s yst em dema ndin g
specific flood control measures (Gyawali 2008; Kale 2008;
Sinha 2008; Dixit 2009). Previous articles by Ajaya Dixit (2009),
Rajeev Sinha (2008) and D K Mishra (2008) as well as other
publications by Kale (2008) and Reddy et al (2008) discuss in
detail the hydro geology of the Kosi river and the flood measures
taken. Therefore, we will provide only a short summary. The
Kosi basin rises from 8,000 m above sea level (ASL) to 95 m ASL
and comprises wide-ranging geological and climatic charac-
teristics (Dixit 2009). The river is highly volatile with variable
discha rge (K ale 200 8; Si nha 200 8). Abo ut 50 % of the Kos i ba sin lie s
4,000 m ASL and only 16% of the Kosi basin lies below 120m ASL.
This implies that there is not enough space to accommodate
the enormous run-off generated by 84% of the Kosi basin
(Kale 2008). The Kosi is one of the heaviest sediment-carrying rivers
(80 million m3/year) due to cloudburst, Asian monsoon, mass
wasting, and Himalayan landslides. The basin is also seismically
active with tectonic movement (Dixit 2004 and 2009; Kale 2008;
Sinha 2008; Reddy et al 2008; Gyawali 2008). The low hydraulic
gradient of the Kosi river, 1:5000m (Reddy et al 2008), cannot
flush these enormous sediments. Instead, the Kosi shifts to a new
course seeking a higher hydraulic gradient (Reddy et al 2008;
Sinha 2008; Kale 2008; Dixit 2009). Thus, flooding either by high
discharge or by shifting of the Kosi river course is a natural
phe no me non t hat rem ain s devast ating to t he people of Bihar.
After severe floods in 1953-54, social and political pressure led
the Indian government to prioritise the issue and negotiate the
Kosi treaty. Subsequently, embankments for flood control were
bu ilt alo ng t he Kosi riv er a lth ough s ome eng ine ers at t he t ime f elt
that given the high sediment load in the Kosi river, embankments
might not be the b est solution (Mi shra 2008b; D ixit 2008).
With the embankments completed in 1959, the sediment load
had nowhere to go and being deposited within the embankments,
over time it increased the height of the river bed by up to four
metres above the surrounding land (Mishra 2008a; Dixit 2008).
Moreover, the morphology of the river changed considerably and
the basin lost its natural drainage system (Dixit 2009). As the
river sought to continue its flow unrestricted, the power of the
water continually put pressure on the weak points in the embank-
ments. Consequently, the embankments breached several times
SPECIAL ARTICLE
Economi c & Politic al Weekly EPW january 9, 2010 vol xlv no 2 47
however, I was not too worried as the embankments were still
there. I never thought it would breach the next day” (Kusaha
v illage, 20 December 2008). Villagers appear to trust the embank-
ments and have little of awareness of the risk of breach under spe-
cific conditions. According to the Nepali senior officer from the
ministry of home affairs, there is no flood warning system in Nepal
nor has flood forecasting been developed (I41, 30 January 2009).
However, there were indications that an embankment breach
was imminent. An army officer from the Kosi Tappu Wildlife re-
serve in Kusaha village, Nepal where a portion of the embankment
lies, informed the local authority about the weathering of the em-
bankment a few days previously (I6, 2008).6 The Kosi project o ffice
in Birpur, Bihar already knew that the Koshi had diverted to the
eastern embankments and that Kusaha point was vulnerable from
the first week of August 2008. However, the chief engineer of the
Kosi project reported that everything was fine until 16 August
2008 (Thakkar 2009: see end note 5). There were news in local
newspapers from Sunsari district about the rapidly weathered
spurs at Kusaha in Nepal on 16 August 2008 (I7, 2008).7 In the
night prior to the embankment breach, local people from Kusaha
village made several calls to the local newspaper warning of the
probable danger (I7, 2008). However, the local authority did not
feel any urgency to evacuate the people to safety.
The embankment breached around 12.55 pm on 18 August
2008. While some fled, others were reluc ta nt to leave t heir inun-
dated houses and belongings for fear of theft (I6, 31 December
2008). The flood victims said: “it was only after the embankment
breach that the news spread like wildfire and everybody was
running here and there. We could not take any of their belongings”
(I4, 25 December 2008). The flood victims were thankful at least
that the breach was not at midnight, which would have greatly
reduced their chances of survival. The lack of flood preparedness
was con firmed by of ficer s fr om the min ist ry of home affa irs , w ho
explained that Nepal was not prepared for, or able to adequately
anticipate, this kind of disaster (I3, 10 January 2009). He pointed
out the reason being a lack of financial resources.
In Bihar, news coverage revealed that flood preparedness and
rescue efforts were poor and that assistance was provided a week
after the disaster took place. Mishra (2008a) also mentions short-
comings in flood preparedness and rescue work in Bihar.
Probable Embankment Breach Not Prioritised: By the first
week of August 2008 and possibly before, the Kosi project office
in Birpur, Bihar which has the responsibility of maintaining the
embankments, already knew that the Kusaha point in Nepal was
vulnerable ( I8, 20088; Thakkar 2009: end note 4). On 16 August
local authorities in Nepal (I61, 2008), the Indian embassy in
Kathmandu (I9, 21 January 2009)9, the liaison office of the state
government of Bihar (GOB) ( I10 , 20
January 2009)10 i n K a t h m a n d u,
and the Ministry of Water Resources in Nepal (I11, 25 January
2009)11 were informed of the matter. The central government’s
ministry of water resources in Nepal instructed local authorities
in Sunsari district to cooperate with the Kosi project office for
maintenance works. However, the newly appointed chief district
officer (CDO) of Sunsari was unaware of this. In a meeting with
the chief engineer from the Kosi project, on 16 August 2008, he,
however, agreed to cooperate with the local autho rity in Nepal.
In addition, local people and an army officer from Kusaha
village in Sunsari district approached the local a uthority of
Nepal several times after 16 August but maintenance work did
not start. From 16 August to the morning of 18 August, there
were more than 48 hours to act on the visible signs of this im-
pending disaster. As only spurs 12.9 and 12.11 were weathered by
the Kosi on 16 August, a concentrated effort could have possibly
averted the embankment breach and the subsequent devastation.
However, the warning from the local people was not heeded, the
danger was underestimated, and time was spent debating in
meetings rather than taking action.
Lack of Monitoring and Maintenance of the Embankments:
According to the Kosi treaty, the water resource department
(WRD) of the government of Bihar is responsible for maintaining
the embankments up to 32 km upstream of the barrage in Nepali
territory. Moreover, the Kosi High Level Committee (KHLC) , u n d e r
t he c ha ir ma ns hi p of th e G an ga F lo od Cont ro l C omm issio n (GFCC)
in Patna is responsible for monitoring the protection work carried
out under the Kosi project. It was the responsibility of the KHLC
to recommend further protection measures to be implemented
before the next flood season (I12, 30 January 2009).12 The two
Nepali members of KHLC should have been informed of, and
invited to, joint monitoring arrangements in Nepali territory.
A ccording to the senior officer of the Nepal ministry of water
r esources, it was agreed at a meeting of the Joint Committee on
Water Resources (JCWR) between Nepal and India in 2001, that
the embankments were defunct and needed strengthening. But
after the 2007 monsoon, there was no communication by KHLC to
the Nepali members (I 12, 30 January 2009) or from the Nepali
members to the KHLC, with the result that no joint monitoring
was undertaken. In this period of time, the priority of the Nepal
government was focused on the peace agreement with the Maoists
and bringing them into the political mainstream. According to
the officer in the liaison office of the Kosi project in Biratnagar,
Ne pa l, so me I nd ian o fficer s c ame for a sit e v isi t in t he Kos i P roj ect
area in Nepal without informing their Nepali counterparts in
advance. Apparently they did not detect any problem and
returned to India. The Nepali members did not investigate
whether the monitoring was done or not. Such a lack of communi-
cation between Nepal and the Kosi project, WRD, GOB demon-
strates a lack of genuine interest in monitoring the e mbankments.
In a report submitted to the Kosi enquiry commission regard-
ing accountability for the Kosi embankment breach of 2008,
Himanshu Thakkar from the South Asia Network argues that
based on evidence, the KHLC did not do their duty from 1 October
2007 until after the end of the monsoon period of 2007. He
further stated that three consecutive letters, dated 1 April 2008,
25 April 2008 and 12 June 2008, were sent by the Ganga Flood
Control Commission (GFCC) to WRD, GOB asking for a r eport on
protection work. However the WRD, GOB did not re spond to t hese
letters. Instead, the chief engineer reported on 16 and 17 August
2008 that the embankments were safe (Thakkar 2009: see end
note 5). By the end of April 2008, the maintenance work should
have been completed before the snow melt of the Himalayas
SPECIAL ARTICLE
january 9, 2010 vol xlv no 2 EPW Economi c & Politic al Weekly
48
reached the Kosi. Maintenance work undertaken in the monsoon
season of August was not effective .
According to literature reviewed and interviews made, main-
tenance and monitoring of the Kosi project was apparently not
prioritised by the Kosi project, WRD, Bihar. Newspaper reviews
(Gyawali 2008: 5), Mishra (2008) and Dixit (2009) mentioned
ongoing corruption in the name of maintenance and monitoring
in the Kosi project. Neither were there systems of accountability
and compliance in place to ensure actual implementation of
e mbankment maintenance activities.
The Nepali government was also negligent in this matter.
Although the maintenance and monitoring of the embankments
in Nepalese territory is India’s responsibility according to the Kosi
treaty, the people living there are Nepalis. The first victims of a
flood would be the Nepali people and their land. One of the s enior
officers from the Nepali Department of Water Induced Disaster
Prevention (DWIDP) explai ned th at:
32 km upst rea m of t he ba rrage is India’s r espon sibi lit y (alb eit i n Nep a-
lese territor y) and GOB has the responsibility and the right to operate
he re. We ca nn ot tou ch t hos e pa rt s an d we d o no t ha ve bu dge t fo r mon -
ito rin g and ma inte nanc e of t hose emba nkm ents(I17, 3 0 January 2009).
Every Nepali government officer interviewed provided a simi-
lar argument and said that the maintenance was neither their re-
sponsibility nor that Nepali authorities prioritise this matter.
However, they do agree that maintaining law and order in the
Kosi project area in Nepal is the responsibility of the Nepali govern-
ment. The Kosi project, Birpur, GOB claimed that they could not do
the maintenance work in the first week of August due to local law
and order problems in the Nepali area. Senior officers from local
as well as the central government of Nepal said that they were not
aware of such problems and that the government of India did not
make any attempt to communicate this with the government of
Nepal. The ministry of home affairs said they were not informed
about the problem of law and order in the project area. Only on
the 16 August did the Indian embassy of Nepal communicate with
the Nepali ministry of home affairs and the ministry of water re-
sources of Nepal about the need for cooperation in maintenance
work in the Kosi project area of Nepal.
The contractors appointed by the Kosi project office, WRD,
GOB came to maintain the embankments in the peak monsoon
period in 2008. According to local people and a local journalist
(I56, 19 December 2008), there was a strike in Sunsari district,
Nepal called by political extremist groups forbidding any trans-
portation and movement. It is understandable that nothing could
have been done during the strike period due to the risk of physi-
cal harassment. In the words of an assistant engineer of the Kosi
project in Birpur, Bihar: “everybody loves their life so nobody
would take that risk and do the work”. The strike was over on the
14 August 2008 leaving three full days and four nights to act
before the embankment breached.
According to the flood victims, the embankments in Nepal
were not maintained for seven or eight years (I14, 20 December
2008). Moreover, corruption among Nepali and Indian
contractors affected the effectiveness of maintenance work
(Interviews, Mishra 2008; Gyawali 2008; Bharati 1997 cited
in Dixit 2009). A junior officer from the Kosi project office
EPW Research Foundation (A UNIT OF SAMEEKSHA TRUST)
C-212 Akurli Industrial Estate, Akurli Road, Kandivli (East), Mumbai - 400 101
Tel: (022) 2885 4995/ 96 & 2887 3038/41 | Fax: (022) 2887 3038 | Email: epwrf@vsnl.com | web: www.epwrf.res.in
A
nnual Survey of Industries
1973-74 to 2003-04 (Vol. II)
A Data Base on the Industrial Sector in India
The EPW Research Foundation (EPWRF) has released the Second Edition of its study on the Annual Survey of Industries
(ASI), thus extending its data base for 31 years from 1973-74 to 2003-04 in CD ROM. This study is one of the most coveted
data base studies of the institution. This is considered useful for wide sections of professionals - research scholars at universi-
ties/colleges/institutes as also policy planners and company executives in India and abroad alike.
We hope that this revised data base work will facilitate more serious research on the nature of industrial growth in India in
recent decades and its various dimensions. The CD containing the publication is priced per copy thus :
CD ROM
Inland Rs. 4,500*
Foreign US $ 200
* Concessional rate of Rs. 2,500 for subscribers of the previous volume; so also for students and research scholars on
producing a brief evidence of their eligibility for concession along with the order.
NB: Some print versions of the First Edition are available for sale at a nominal price of Rs. 800 per copy.
CDs (and First Edition copies) can be obtained from:
The Director, EPW Research Foundation
C-212, Akurli Industrial Estate, Akurli Road, Kandivli (East), Mumbai – 400 101, INDIA.
Phones: (022) 2885 4995/ 4996 E-Mail: epwrf@vsnl.com
(Remittances by Demand Draft/Cheque or through online payment gateway. Demand Draft and Cheque are to be in favour of EPW Research Foundation payable at
Mumbai. Outstation Cheques should include Rs 50/- as bank collection charges. For online payment, please visit our website: http://www.epwrf.res.in)
Second Edition
SPECIAL ARTICLE
Economi c & Politic al Weekly EPW january 9, 2010 vol xlv no 2 49
(I51, 24 December 2008) alleged that less than half the mainte-
nance budget is actually spent.
The consequences of the negligence described above are that
there are 3.5 million flood victims from Bihar and 45,000 from
Nepal. The misery and trauma they are facing today could possi-
bly have been avoided if things were done honestly and on time.
Moreover, this was not the first embankment breach; the past
history of embankment breaches should have made officials
anticipate possible breaches and act more effectively (Mishra
2008a). Although the disaster was foreseeable, the embankments
were allowed to weaken. Not only do the people living on the
banks of the Kosi have to deal with the dynamic hydrogeology of
the Kosi river (especially the sediment loads), they are also not
protected by the institutions managing this river. While the
former needs to be subject to high quality scientific and technical
analysis, the latter calls for institutional analysis.
Complex Communication Mechanisms and Gaps of the Kosi
Project: Nepal and India signed the Kosi treaty in 1954 for the
primar y purpose of flood control along the Kosi by building
embankments under the Kosi project, although irrigation and
hydropower production were included in the treaty. The treaty
was signed by the government of India (GOI) and government of
Nepal (GON). However, the authority for the execution of the Kosi
project was given to the WRD, GOB. The GOI does not interfere in
the functioning of the Kosi project as it is a state matter of Bihar
(I12, 30 J an uar y 2 009). When t he re are i ss ues t o be d is cu ssed wi th
Nepal, the WRD seeks guidance from the GOI (I9, 21 January
2009). There is no involvement of the Nepali govern ment in the
Kosi project other than one liaison and land acquisition office in
Biratnagar, Nepal. The office is run by Nepalese staff appointed
by the Nepali ministry of water resources and financed by the
Kosi project, WRD, GOB.
Chart 1 shows the communication mechanism and the manage-
ment organo-gram for the Kosi project in WRD, GOB. The lines
of communication can be described as lengthy, indirect and
complex. We identify two major problems: (1) bureaucratic hier-
archy, and (2) trans-boundary sensitivities. With regard to bu-
reaucratic hierarchy, when action in the field is needed, field-level
officers do not have sufficient authority to act without seeking
approval further up the management hierarchy (I8, 24 December
2008). As shown in Figure 1, the communication from the junior
officer to the decision-making authority is lengthy and has to pass
through several levels of authorisation before it reaches the WRD
secretary where a decision can be taken. This obviously entails
a huge lo ss of ti me.
Trans-boundary communication between Nepal and India suf-
fers from discontinuity. The maintenance of the embankments in
Nepal is the responsibility of the WRD in Bihar. None of the local
Nepalese officers have the authority to directly communicate
with the WRD, the GOB or GON. But the WRD, as part of GOB, also
does not have any direct communication mechanisms with the
GON. The liaison office of the Kosi project in Nepal organises the
communication between the WRD and the local and central
authority in Nepal. However, the local government of Sunsari
district in Nepal is not directly involved in the communication
process for the Kosi project despite the eastern embankments of
the Kosi river in Nepalese territory being located in this district.
This is because the local government falls under the ministry of
local government (MLG), an d the K osi pr oject fal ls u nder t he m in-
istry of water resource’s jurisdiction. The ministr y of water re-
sources must communicate with MLG that, in tu rn, must commu-
nicate with local government. The 2008 Kosi flood has exposed
th e c onti nuing cha llenges in t he commu nic at ion processes; t hese
channels are complex, slow and ineffective.
Furthermore, field research indicates that the messages com-
municated may be incorrect or incomplete. The chief engineer of
the Kosi project, WRD, GOB gave the impression that the emba nk-
ments were under control as of 16 August 2008 (Thakkar 2009:
see endnote). The GOI was not aware of any problems along the
Kosi in the Nepali territory because communication from both
the field-level officers in Bihar as well as Nepal did not reach it in
time. Neither did the Kosi project office, WRD, GOB seek coopera-
tion from the liaison office in Nepal nor from the central govern-
ment from Nepal and India until it was too late. They might have
underestimated the probable disaster. Thus, both the GON and
GOI were unaware of the deteriorated conditions of the embank-
ments. It is known that local people and the army camp informed
the chief district officer in Nepal of the worrisome state of em-
bank ments. It was only on 15 August, that the WRD informed the
Indian embassy in Kathmandu of the situation and requested full
cooperation from Nepal.
Liaison office based in
Project Area in Nepal
Indian Embassy in
Nepal
GOB
Water Resource Department
(Secretary)
Administration Head
GOI
Secretary of water
Resource
Technical Head (Engine er in Chief)
Zone Head (Chief Engineer)
Circle Head (Superintendent Engineer)
Division Head (Engineer)
Junior Engineer)
Ministry of Water
Resource, Nepal
MLD
Local people in Nepal /Army
Camp, Kosi Tappu wild life
reserve
Local government
Nepal
Department of
Irrigation, Nepal
Ministry of Foreign
Affairs
Sub Division Officer
(Assistant Engineer)
Chart 1: Communication Mechanism for the Management of the Kosi Project within
WRD, GOB and between GOI and GON and Gaps in Communicatio n before 2008
Embankment Breach
Sou rce: Fiel d res ear ch 20 08- 09. T he d ott ed li ne me ans i nst itu tio nal h iera rch y in Ne pal ese side as
well as Ne pal and I ndia . Wh erea s the non -do tte d lin e mea ns th e ins tit uti ona l hie rarc hy in Biha r
government and Indian government.
GOI
Secret ary of Water
Resource
Subdivision Officer
(Assist ant Engineer)
Junior Engineer
Local people in Nepal/Army
Cam p, Ko si Tap pu Wi ldl ife
Reserve
GOB Indian Embassy
in Ne pal
Water Resour ce
Department (Secretary)
Ministry of
Foreign Affairs
Administration Head
SPECIAL ARTICLE
january 9, 2010 vol xlv no 2 EPW Economi c & Politic al Weekly
50
Complex and Exclusive Institutional Design of the Kosi Project:
The Kosi project was designed and proposed by the GOI but man-
aged by the WRD of the GOB. The barrage and embankments are in
Nepalese territories; however, their maintenance and monitoring
is managed by the Kosi project as per the Kosi treaty. The Kosi
treaty clearly states that Nepal as a sovereign country should be
consulted by GOI and GON and informed before taking any action
in Nepalese territory. The GON currently has no management in-
fluence on this project. The institutional design allows Nepal to
avoid taking responsibility for important activities in its own ter-
ritory despite the moral imperative for the sovereign government
of Nepal to ensure the well-being of her people.
A furt her level of complexit y is t hat t he Kosi treat y is sig ned by
both GON and GOI, while the responsibility for execution of the
Kosi project is assigned to the WRD, GOB. There is no direct link
between the province of Bihar and the national authority of N epal
at the same political level. Therefore, despite having full control
over operations including emergency work in Nepalese territory,
WRD Bihar must first consult the GOI, which then consults the
GON before undertaking any infrastructure maintenance. This
institutional mechanism is problematic and hinders immediate
decision-making.
Fo r t he wa te r s ec to r t he tw o c ou nt ri es ar e i ns ti tu ti on al ly li n ke d
through the joint committee on water resource (JCWR) between
Nepal and India that oversees several other projects and a number
of rivers including the Kosi. The two main committees related to
the management of the Kosi project are shown in Chart 2. These
include the Joint Committee on Kosi and Gandak Projects and
the Kos i H igh Le vel Comm ittee (KHLC).
Local governments of both countries are not represented, nor
do they have direct access to these committees. They are not
i nformed about the decisions and actions authorised by these
committees (I81, 24 December 2008). A lso communication be-
tween the different offices is not very smooth. For example, as
the KHLC is responsible for monitoring and making recommen-
dations for the maintenance of the embankments along the
Kosi, it tried to assess the maintenance from the Kosi project.
Despite continuous correspondence to the WRD the KHLC ob-
tained no r eply (Thakkar 2009 and I6, 31 December 2008) for
reasons not known. It seems that the central committees have
little information from the local or state level on the functioning
of the projects. This results in a lack of coordination, accounta-
bility, communi cation or ownership of the project leading to a
lack of timely a ction.
The evidence indicates that there are too many bilateral com-
mittees between Nepal and India under the JCWR. The JCWR meet-
ing, after the 2008 flood, acknowledged the complexity of having
too many committees and their ineffectiveness (I12, 28 January
2009). There has been a modification of the institutional mecha-
nism between Nepal and India for managing water resources as
illustrated in Chart 2. However, there is not much difference to the
previous institutional design (I4, 30 January 2009). As shown in
Chart 2, one ministerial level bilateral committee is proposed
above the JCWR and a joint technical standing committee (JSTC) is
formed below JCWR. This has been described as a three-tier insti-
tutional mechanism between Nepal and India on water resources
(minute of JCWR on 29 September 2008). It was felt that issues
could not get full attention because of the lack of authority for
making decisions and implementing them in JCWR level (I4, 20
January 2009; I11, 25 January 2009 and I121, 30 January 2009).
Therefore for discussing issues on water resources between Nepal
and India at a higher level, a joint India-Nepal ministerial commis-
sion on water resource was proposed at the meeting of JCWR on
29 September 2008. This proposed body would bring the issues
to the national parliaments for extensive discussion in order to
seek solutions through political consensus. Similarly, JSTC is
formed bet wee n tw o count rie s at a t ech nic al l eve l. A ll t he techn i-
cal issues will be discussed through this committee. JSTC h as the
responsibility to assess the tasks of the other subcommittees
between Nepal and India on water resources. In the modifica-
tion of the institutional framework after the 2008 floods, it was
proposed to cut the committees and merge them for better func-
tioning (Minute of JCWR, 29 September 2008). However, the new
institutional mechan ism is also centralised a nd there is no inclu-
sion of local government, local people and diverse stakeholders.
Thus, t he in stitution al mec hanism i s not inclusive.
Interviews and the minutes of the JCWR mee ting of 29 Septe m-
ber 2008, indicate there are gaps in institutional functioning.
The gaps are lack of decision-making power and full authority
for the senior officers of the Kosi project office, lack of responsi-
bility and accountability of the Kosi project office and gaps in su-
pervision of the tasks carried out by the Kosi project office. In
addition, there is no mechanism of penalty and compliance for
the dereliction of the duties. These gaps are discussed here.
The head of the Kosi project lacks the authority for giving a “go
ahead” for the maintenance and monitoring of the embankments
(I8, 24 December 2009 and I12, 30 Januar y 2009). It is very difficult
Nepal India
Joint Nepal India Ministerial Commission on
Water Resource (JNIMCWR)
Thi s com miss ion i s pro pos ed a fte r 200 8 fl ood
Joi nt Co mmit tee on Wa ter R eso urce ( JC WR)
Joint Standing Technical Committee (JCWR)
New ly fo rme d com mit tee aft er 20 08 f loo d
Kosi High Level
Committee
(KHLC)
Joi nt Co mmit tee on Ko si
and Gandak Porjects
(JCKGP)
The t eam l ead er fr om In dia
is no w se cret ar y of W RD,
Bihar instead of engineer in
chi efc o f the Kosi proj ec t
Joint
Committee on
Inundation
Joi nt Co mmit tee on
Flo od Co ntr ol an d
Management (KHLC)
Did n ot d o
monitoring since
2006 Monsoon
Did n ot m eet o r
communicate since
Jun e 200 6
Sou rce: Fiel d Res ear ch 20 08- 09.
Chart 2: Institutional Framework between Nepal and India for Managing
the Kosi Rive r an d Gap s in Inst itu tio nal Func tio n
Did n ot m eet o r
communicate
sinc e 20 04
SPECIAL ARTICLE
Economi c & Politic al Weekly EPW january 9, 2010 vol xlv no 2 51
to get a “go ahead” from the monitoring and maintenance
dep artment ( I8, 2 4 January 2009).
Since 2004, there has been no meeting of the JCWR which
should have been conducted every year. Thus there was no
sharing of the problems, issues and progress of any projects be-
tween Nepal and India and hence neither of central govern-
ments was aware of the issues. A senior officer of the ministry of
water resources and department of irrigation said that India did
not want to have any bilateral talks after the king of Nepal took
over the country in 2004. A senior officer from the Indian
embassy said there was no proper government in Nepal to have
bilateral talks with. However, this is not a credible reason, since
there was a government representative and the Indian govern-
ment had c ordial relationships with the Nepali government
(Dixit 2009).
Similarly, JCKGP had not met since June 2006. Thus the issues
regarding the Kosi project which should have been discussed at
least annually were not communicated for nearly two years
b efore the embankment breached on 18 August 2008.
The KHLC under WRD, GOB chaired by the chairman of GFCC
was responsible for monitoring the damage caused by the mon-
soon and recommending any necessary works. However, KHLC
ha d n ot d on e thi s si nc e t he m on soon of 200 6 ( I11, 25 Ja nua ry 2 00 9
and I12, 30 Ja nuary 20 09). T hi s is a lso no ted by T ha kk ar (2 009) in
his letter to the Kosi Enquiry Commission. However there is no
mechanism of penalt ies and compliance for not doing complying
with the prescribed duties. Therefore, there is clear gap in the
functioning of the institutional design in place to manage the
Kosi ri ve r and t he pr oj ect.
The points made above are an indication of the institutional
dysfunction in managing the Kosi river for flood control. How-
ever, this dysfunction is a result of institutional challenges that
wi ll be e laborated below.
Major Institutional Challenges
Trans-boundar y Politics between Nepal and India: All t he riv-
ers from the Himalayas and the hills of Nepal cross the border
and ent er I ndia . Ind ia is an em erging pow er i n sou th Asi a and h as
a major influence on the political environment of Nepal. India
has been successful in claiming its share from all the major rivers
of Nepal under a number of treaties (Dixit 2008). Joint projects
are under construction and included in plans, however, Nepal’s
financial dependence on India has prevented Nepal from being
an equal partner in sharing the benefits from its rivers. India’s
main interest in the Himalayan rivers from Nepal is for irrigation
of the whole north belt of India while Nepal wants to export
hydroelectric power to India. At the same time, Nepalese people
face up to 16 hours of load shedding in the winter and in the mon-
soon. Moreover, water analysts in Nepal argued that India is not
desperate for hydropower from Nepal confirming that irrigation
is the main interest. Flood control has always been the catalyst
for signing treaties; however, it can be argued that flood control
is secondary to the main interest. The irrigation department leads
the joint projects on a number of different rivers, embankments
for flood control are secondary to this purpose.
Internal National Politics of Nepal, Bihar and India: Many
changes and challenges have occurred in the internal politics of
Nepal in the last 15 years. Law and order has suffered from the
Maoist insurgency, the king’s takeover and peace process and the
transformation of the kingdom to a republic in 2008. In the pe-
riod during the peace process, there was no authoritative govern-
ment. The country had just become more peaceful after the Mao-
ists signed the peace accord in 2007. However, this was marred
by the presence of different armed powers in the Terai demand-
ing an autonomous Terai. This has not just affected the peace of
mind of Nepali people but also ever y sector of life including the
maintenance of the Kosi embankments.
Alleged bribery and corruption (Gyawali 2008, Bharati 1997
cited in Dixit 2009 and Mishra 2008a)13, confirmed by field
research, has played a role in the Kosi project for half a century.
This has reduced the effectiveness of flood management proc-
esses and has also led to a politicisation of the issue where local
labour difficu lties and strikes are cited as the cause for not being
able to maintain the embankments.
With regard to India, the nature of the relationship between
GOI and GOB is be yon d the foc us of this p ape r bu t may be rel evant
to mana gem ent of the Kos i ri ver. I n th is sc ena rio, trans -bo und ary
relationships are not easy. The Kosi river and project have
s uffered from the internal political environment in Nepal and the
intr ast ate p ol it ical re lationsh ip bet we en Bih ar and In dia.
Wea knes s of t he Kos i Trea ty
The Kosi treaty, originally designed to help control flooding of
the Kosi river contains no specific clause for flood control mecha-
nisms or compliance in cases where the flood control structures
either fail or are damaged. The institutional mechanism of the
Kosi projec t does not appor tion a ny re spon si bility for the mainte-
nance and monitoring on the Nepali side of the river even though
some parts of the embankments are in Nepali territory. There-
fore, when the embankments breached in Nepal, it proved
d ifficult for India to provide timely action to repair the damage in
Nepali territory as there were no provisions for such measures in
the Kosi treaty. The issues of jurisdiction of land and sovereignty
were simply not address ed.
There is no clear description in the treaty for dealing with a
flood when it does occur. There were no contingency plans in
place should the embankment breach. There is merely a mainte-
nance plan for the embankments, the responsibility for which fell
to India. Nepal had the responsibility of keeping law and order in
the project area under her territorial jurisdiction. Given that a
breach was not anticipated there were no compensation schemes
for the people who would be affected by the flood if the embank-
ment was breached. In short, the possibility of the occurrence of
a future flood was not addressed. Respondents opined that the
overconfidence over the output of the embankments is one of the
many reasons. Moreover, there are still no policies and strategies
legislated in Nepal for disaster or risk reduction which the Nepal
government claimed that it is working on it (Prime Minister of
Nepal, 15 October 2009, Himalayan Times, p 2).
The Kosi treaty was signed by Nepal and India and decisions
were made at a national level. Local governments, local people
SPECIAL ARTICLE
january 9, 2010 vol xlv no 2 EPW Economi c & Politic al Weekly
52
and civil society were not included in the process of decision-
making or in the management of the Kosi project; nor are they
included in the new institutional design. Neither the local people
(I14, 1-30 December 2008),14 nor the local government and com-
munity based organisations are aware of the detail of the Kosi
treaty (I 15, 18 December 2008).15
Nee d for Revi ew of t he Ko si Tre aty
Since some of the authority to take action comes from the Kosi
treaty, this paper suggests that it would be appropriate to review
the Kosi treaty. Following points need to be considered.
Goals of the Kosi Treaty: The Kosi treaty was born with the
main objective of flood control along the Kosi. However irrigation
became prime focus for both of the countries and irrigation de-
partments lead the Kosi projects. There are no institutions
e specially set for disaster preparedness under the Kosi treaty and
the Kosi project. Thus it needs to be internalised inside the Kosi
treaty for th e K osi rive r.
Monitoring: The treaty needs to be amended to show clearly
when, who and how monitoring would be conducted. The divi-
sion of costs for maintenance, monitoring and accountability pro-
cedures to ensure that these budgets are effectively spent is
needed. In addition, it is important to define clear roles and re-
sponsibilities for the local authorities and people when drafting
such strategies. These actors would feel a sense of ownership of
the embankments and understand their importance. This owner-
ship and awareness might lead to better on-site monitoring as it
would not call for long distance visits and may even reduce the
cost of the monitoring carried out by the state government. In
a ddition, those embankments which are in Nepali territory could
be monitored by Nepal, thus creating greater accountability and
will remove the complications of India working in Nepali
s overeign territory.
Maintenance: The treaty could be amended to clearly explain
the responsibility for every aspect of maintenance. It would help
if the treaty is clear about when and how this work is to be car-
ried out. As Nepal is the first victim of any disaster upstream of
the barrage, it would be easier for the management of the Kosi
project if Nepal would be given responsibility for maintenance of
embankments that are in Nepali territory. Funding arrangements
will have to be reviewed and money to carry out the above tasks
reallocated appropriately. However, there needs to be much more
intensive cooperation between those undertaking maintenance
activities in both countries.
Accountability and Transparency: The treaty needs to be
amended to clearly define accountability for performing specific
tasks and responsibilities. As there is strong suspicion of
c orruption in the Kosi project, there is the need for clear systems
for financial accountability. Institutions would be more responsi-
ble for the assigned tasks if there would be clear penalties en-
forceable by national and/or international law for those who do
not adhere to them.
Compensation: The mechanism for compensation of flood
victims needs to be addressed and schemes to generate
finance to enable this could be defined and implemented. Such
schemes could include insurance schemes, micro-financing
and so on.
SAMEEKSHA TRUST BOOKS
1857
Essays from Economic and Political Weekly
A compilation of essays that were first published in the EPW in a spe cial issue in May 2007. Held together with an introduction by S ekhar Bandyopadhyay,
the essays – that range in theme and subject from historiography and military engagements, to the dalit viranganas idealised in traditional songs and the
“unconventional protagonists” in mutiny novels – converge on one common goal: to enrich the existing national debates on the 1857 Uprising.
The volume has 18 essays by well known historians who include Biswamoy Pati, Dipesh Chakrabarty, Peter Robb and Michael Fisher. The articles are grouped
under five sections: ‘Then and Now’, ‘Sepoys and Soldiers’, ‘The Margins’, ‘Fictional Representations’ and ‘The Arts and 1857’.
Pp viii + 364 2008 Rs 295
Available from
Orient Blackswan Pvt Ltd
Mumbai Chennai New Delhi Kolkata Bangalore Bhubaneshwar Ernakulam Guwahati Jaipur Lucknow
Patna Chandigarh Hyderabad
Contact: info@orientblackswan.com
SPECIAL ARTICLE
Economi c & Politic al Weekly EPW january 9, 2010 vol xlv no 2 53
Notes
1 Interview given by Deepak Gyawali to reputed
nationa l daily of Nepal (The Kathmandu Post,
1 September 2008, p 5). Gyawali is the former
minister of water resource in Nepal and founder
of Nepal water conservat ion foundation.
2 UN OCH A is hosting Gene ral Coordinat ion Meet-
ing (GCM) among different relief organisations,
local governments and donors since the embank-
ment breach. In thi s meeting stake holders meet
and discuss issues and gaps in relief activities and
st atus of floo d vi ctim s.
3 To keep the respondents anonymous we only pro-
vi de co de and date o f the int erv iew.
4 Intervie w held on 30 Januar y 2009 w ith a senior
officer from Department of Hydrology and Meteoro-
logy, N e pal. He is also the member of JCKGP.
5 Submission to the Kosi Enquir y Commission on
Ko si emb ank ment brea ch 20 08 (10 Ma rch 2009).
6 Inter view (31 December 2008) wit h a senior
officer, Sunsari dist rict office, Sunsar i, N epal. He
is also the chair person of District Disaster Relief
Committee.
7 Interview (17 December 2008) with a local journal-
ist and a com munit y-based organisation. The
jo urn ali st is bas ed i n Sun sar i di str ict a nd h e is t he
witnes s of ongoing events before and after the
Kosi embank ment breach 2008. U PCA Nepal and
Abhiya n N ep al are t wo communit y-based o rga ni-
sations interviewed for this research.
8 Interview (24 December 2008) with an engineer,
Kosi project Birpur, Bihar, India. He was inter-
viewed during the closure of the Kosi embank-
me nt bre ach 2 008.
9 Interview with a senior officer from Indian Embassy,
Kathmandu, Nepal.
10 Inte rview with the li aison officer, WRD, GOB,
Kathmandu, Nepal.
11 Interview with the Joint Secretar y, Mini stry of
Water Resource, Nepal.
12 I nterv iew with a sen ior of ficer from Depar tment
of Irrigation, Nepal. He is a lso the member of
KH LC a nd tea m lea der of JCKG P fro m Nepa l.
13 See D K Mishra’s “Serp ent Tied Around Neck: Kosi
Emb ankm ent”. Here Mishr a has l isted a ll th e pre-
vious emba nkments breac hed and the ci rcum-
st ance s b e fore a nd du rin g the breach es.
14 I nterv iew w ith flood v icti ms in Kosi flood-a ffec t-
ed areas in Nepa l. Nearly 20 flood victims were
in terv iewe d fr om th ree a ffe cted vil lage s.
15 I nterviews with community-based organisations
an d loc al gov ern ment o ffi cers .
References
Dixit, A (2004): “Floods and Vulnerability: Need to
Rethink Flood Management”, Natural Hazards,
28, 155-79.
– (2008): “Two Neighbours Water Course” (Nepali
language), (Kathmandu, N epal: Action Aid Nepal).
– (2009): “Kosi Emba nkment Breach i n N epal:
Need for a Parad igm Shift in Resp onding
to Floods”, Eco nomic & Politica l Weekly,
7 February.
Gyawali, D (2008): “Wh ither Kosi R iver?”, The Kathn
Post, 1 S epte mber, p 5.
Kale, V S (2008): “Him alaya n C atast rophe that
E ngulfed Nort h Bi har”, Journal of t he G eolog ical
Society of India, 7: 7 13-19 .
Mis hra, D K (2008a): “Se rpent T ied Aro und the Nec k:
Kosi Embankment” ( Hindi lang uage), ( Patna,
In dia: F reed om fr om Fl oods Camp aig n).
– (2008b): “Bihar Floods: The Inev itable Has Hap-
pened”, Economic & Political Weekly, 6 September.
– (2008c): “The Kosi and Embankment S tory”, Eco-
nomic & Political Weekly, 15 September.
Reddy, D V, D Kumar, D Saha and M K Mandal
(2008): “The 18 August 2008 Kosi Embankment
Breach: An Eva luation”, Cur rent Science, 9 5(12):
1668- 69.
Si nha , R (2 008) : “K osi: Ri sin g Wat er s, D yna mi c Ch an-
nels and Human Disasters”, Economic & Political
Weekly, 15 November.
Thak kar, Hima nshu (2009): “Submission to t he K osi
En qui ry Co o n Ko si E mba nkm ent Bre ach 200 8 (10
March 2009)” by the South Asian Network, Rivers
and People.
Institutional Arrangements: The process of institut ional design of
flood management in the treaty could include multi-stakeholder
participation across different levels. It could consider the partici-
pation of local people, local government, community-based or-
ganisations; interdisciplinary academics, state government and
central governments. This could be achieved by establishing a
river basin commission made up of these stakeholders. Clear de-
cision-making procedures should be set up and better multi-level
governance systems.
At present there are many bilateral committees in the middle
of the hierarchy for the management of the same river. This has
caused an overlap of responsibility for many tasks and the output
of these committees has not been satisfactory. A streamlined
d ivision of responsibility and authority to actors at different lev-
els of governance, and emergency procedures that can bypass
hierarchical lines of communication are recommended. How-
ever, this does not imply that specific committees cannot be
e stablished with clearly delineated roles and reporting tasks. A
committee is needed with an overall coordinating function but
the power to take action on defined tasks would remain within
these “task oriented” committees.
Local authorities from Nepal and Bihar could thus be empow-
ered to take decisions and implement them at a time of urgency.
Communication Mechanism: The current communication
mechanism is indirect, lengthy and unaccountable. The treaty
could be amended to allow for more direct communication from
the Kosi project, WRD, GOB with the GoN at both the local and
central levels. In addition, more direct communication from the
central level of Nepali government to the local government needs
to be less bureaucratic in times of urgency. In the 55 years that
have passed since the Kosi treaty was drawn up, there has been
major progress made in communication technology and this
progress should be taken into account in the redesign process.
Conclusions
This paper has argued that the disastrous 2008 flood on the Kosi
river calls for an examination of what went wrong and how can
such a flood be prevented in the future. It focuses only on the
i nst it uti ona l a spe ct s an d de mon st rat es t hat t he ser iousn ess o f the
flood can be attributed to five direct reasons, namely, the lack of
local awareness and disaster preparedness, lack of a nticipation
and prioritisation given to the possibility of an embankment
breach, lack of monitoring and maintenance of the embank-
ments, the hierarchical communication mechanism, and the ex-
clus ive and comple x natur e of th e in stitutiona l des ign for deali ng
with the Kosi. These reasons can be further attributed to three
higher-level challenges. These include the nature of transbound-
ary politics between Nepal and India, the internal national poli-
tics of Nepal, Bihar and India, and the inherent weaknesses of the
now-outdated Kosi treaty.
The paper suggests that some of these challenges can be
addressed by making the goals of the Kosi treaty more compre-
hensive, by ensuring a clear division of responsibility between
the two national governments and the local governments, and
a ccountability mechanisms to ensure that activities listed are im-
plemented. A new feature would be compensation mechanisms.
The institutional arrangements should be redesigned and stream-
lined to allow for better decision-making processes and commu-
nication processes.
A technical solution such as embankments has proved to have
exacerbated flooding along the Kosi river. The 2008 flood is a
wake-up call highlighting the urgent need to find alternative so-
lutions to the flood problems of the Kosi. Multi-stakeholder par-
ticipation involving local people and local government in the
process of decision-making and implementation is one way
forward. In addition, clear tasks, more direct communication,
accountability and enforcement need to be prioritised for ensur-
ing a ll affected are re sponsi ble for the matter s of t he Kosi.
