Alliance One Tobacco (Malawi)
Academic Research Repository at the Institute of Developing EconomiesNot a member yet
52305 research outputs found
Sort by
Architecture of Elsewhere: A Center for Accountability at the Site of Extraction
No full textContemporary building and design practices suffer from an inherent detachment
from the harmful extractive processes that sustain them. To transition
away from a carbon-based economy, architecture must become an
active agent in narrowing the gap between itself and these hidden externalities
elsewhere. The Center for Accountability seeks to address the issue
of elsewhere by offering a new model for hyper-local mine governance in
Minnesota’s Arrowhead Region, where several harmful copper and nickel
mines await regulatory approval. It functions as a regulatory facility that
incentivises best practices, centers local and indigenous knowledge, and
challenges the neoliberal underpinnings of contemporary extraction.
The center not only employs a Federal Land Trust model as its conceptual
framework, but it architecturalizes the many regulatory and supporting
bodies within it in functional and perceptual ways. Regulatory bodies shift,
rotate and float around the user, who is given open access to the facility
through a pedestrian circuit. Visitors may visit the Center to protest mine
operations, collaborate with activists and scientists, or simply eat a meal
alongside mine workers, all while bearing witness to the transforming
landscape around them.
The Center’s largest components - its structure, cladding and foundation
- reuse elements from the abundant shuttered mine facilities that dot its immediate
site, allowing the building’s form to act as a floating datum against
which the landscape might be measured.Department of Architectur
Disambiguating Large Language Model Performance on the Ambiguities of Law, Reasoning, and the Future
No full textOver the past year, two separate cases of legal actors using LLM-based products to submit fictitious materials to American courtrooms have raised significant concern. In this thesis, I tackle one facet of this legal LLM usage problem, in arguing normatively for the need to teach LLMs legal reasoning. Drawing from legal theory but developing new definitions for the LLM context, I establish that legal reasoning requires reasoning over a central ambiguity in the application of general statutes to specific fact patterns — an inescapable ambiguity that I find traditional LLM reasoning research paradigms unequipped to handle. Thus, I contribute back to LLM reasoning research an experimental methodology that directly targets this central ambiguity and intuition in reasoning, as well as an evaluative framework that holistically captures true reasoning ability. I rework two prior legal ML datasets into benchmarks for a novel legal ambiguity identification task (available on HuggingFace). Conducting a preliminary experimental exploration of GPT-3.5 and GPT-4 models on this task, I discover some true ability at identifying legal ambiguity. Though it is currently weak and somewhat misaligned with legal practice, this ability shows promise for improvement, especially through model milestoning and fine-tuning. Overall, this thesis analyzes one component of ideal legal LLM usage from an interdisciplinary legal theory and computational perspective, toward progress in principled legal LLM implementations and more productive law and CS collaboration.Computer Scienc
Regenerative Farming From the Margins in Andhra Pradesh, India: An Analysis of Policy and the Intersectionality of Caste and Gender
No full textRegenerative agriculture is emerging around the world as an alternative to environmentally degrading conventional farming practices, while producing enough food to feed the ever-growing world population (Lemke et al., 2024; McLennon et al., 2021). However, farmers marginalized by race, caste, or gender face barriers in shifting to regenerative agriculture due to factors such as limited access to information and credit (Day & Cramer, 2022; Karthick & Madheswaran, 2018).
Caste and gender oppression are openly practiced across India, including in the agriculture sector in the state of Andhra Pradesh (Teltumbde, 2010). In Andhra Pradesh, the average landholding of farmers from the Scheduled Caste (SC), Backward Caste (BC), and female farmers is less than two ha, while the average landholding of the Other Caste (OC) male farmers is more than 10 ha (Department of Agriculture, 2021). Since issuing the Swaminathan Report in 2006 that made several agriculture-related policy recommendations, the Government of India and the state government of Andhra Pradesh have enacted public policies to support farmers and promote regenerative agriculture (PRS Legislative Research, n.d.; Ministry of Agriculture & Farmers Welfare [MAFW], 2024). Assessing gaps that still exist would help in designing policies that support the well-being of all farmers as they transition to regenerative farming (Chopra, 2019).
My research aimed to examine how policies that promote regenerative farming in Andhra Pradesh are accessible to marginalized farmers, and how they improve their socioeconomic conditions. To address this, my research tested the following hypotheses: 1) between 2006-2024, a higher percentage of male OC farmers shifted to regenerative farming compared to other farmers; 2) between 2006-2024, a higher percentage of male OC farmers accessed public policies compared to other farmers; and 3) public policies that exclusively support marginalized farmers improved their socioeconomic conditions.
To test these hypotheses, I developed a survey instrument to collect data on the socioeconomic conditions of the farmers in Miduturu and Pagidyala villages in Nandyal District. Statistical analysis was conducted by creating various cross-tabulations of caste-gender categories of farmers in the two villages, comparing them against their socioeconomic conditions, as well as their accessibility to and use of public policies.
The survey revealed that compared to other categories, OC male farmers shifted to regenerative agriculture at a lower percentage or with an insignificant percent difference. A similar result was found for the accessibility of policies, with OC male farmers accessing policies at a rate not significantly different from other categories. The study found that Jagananna Jeeva Kranthi (JJK) scheme is the only scheme exclusively available to SC and BC female farmers, but it was not adequately accessed, mainly due to a lack of awareness about the scheme, with caste and gender biases reported by few farmers. However, shifting to regenerative farming led to improvement in the socioeconomic conditions of the farmers.
Farmers identified support for inputs and labor required for regenerative agricultural practices, as well as marketing support, as key policy changes needed to facilitate their shift to or continuation of regenerative agricultural practice. Policymakers need to create opportunities to educate the farmers on the availability and applicability of policies, as this would motivate them to shift to regenerative agriculture.Extension Studie
Biggest Bankruptcy Ever!: Viewing a Drama Written in Verse about Investment Banking Through the Lens of Behavioral Economics
No full textThis thesis furthers the growing area of research of examining literature while considering key topics from the fields of economics and business in an effort to increase understanding of literature. By viewing The Lehman Trilogy, a play by Stefano Massini and adapted by Ben Power, through the lens of behavioral economics, insight can be gained to better understand the characters, their economic decision-making processes, and ultimately their humanity. Considering concepts of behavioral economics provides greater acuity into the story. Throughout the play, various characters make decisions along a 164-year timeline from 1844 when the first Lehman brother arrives in New York until September 15, 2008, when the storied global investment bank collapses with more than $600 billion in debt in the largest corporate bankruptcy in history. By drawing subject matter from behavioral economics, primarily from the works of Daniel Kahneman and Amos Tversky, detailed in Kahneman’s widely acclaimed Thinking, Fast and Slow, deeper insight can be gained
GAIA
No full textWhat is it like being bound to two places? Knowing that you must sustain both or
lose yourself? That is the feeling that Anthia, a young woman and Communer, must
contend with as she transitions into a Scavenger. Anthia hails from a universe where
planets are sentient and rely on their inhabitants to provide stability for them. These
planets bond to individuals known as Communers, entrusting them to speak on their
behalf and communicate their needs. These missives are passed to Scavengers, entrusted
by their home planet of Gaia to search the reaches of space and gather the resources
needed. But not everyone is appreciative of this connection. Others might even kill to get
it! When a pirate attack forces an emergency jump, Anthia and her team wind up stranded
on a foreign world. A foreign world that echoes hauntingly of home. To top it all off,
Anthia and her newfound companions find that everything is not quite shipshape on
board and threats exist within and withheld. Can they figure out why malignant forces
revived an ancient plague in time? And will Anthia choose to accept a bond to a place not
quite her own?
Gaia is a science fiction novel I wrote as a desire to answer that question for
myself. I am also from two places. I am fortunate in many ways because my parents
insisted on maintaining strong roots with our heritage and language. My hope is that by
presenting the experience of searching for connections through a sympathetic character,
other young adults might be able to find the answer for themselves.Extension Studie
Leveraging Border Carbon Adjustments for Climate Finance: Matching Carbon Tax Assets with Carbon Tax Liabilities
No full textThis white paper offers a policy proposal for how border carbon adjustments can increase global climate finance, with a focus on the EU Carbon Border Adjustment Mechanism (CBAM). This proposal is designed to offer a “carrot” to developing countries in the form of new funding for decarbonization, to complement the “stick” of carbon pricing introduced by CBAM. By offering a practical way forward for the CBAM to increase finance for clean energy and other carbon mitigation projects in developing countries, we hope to catalyze large volumes of new investment in the Global South, and also to re-energize support for the CBAM as a vital measure in the energy transition.
At its core, CBAM is designed to preserve the level playing field of European firms subject to the EU ETS by imposing an equivalent carbon price on foreign imports. At the same time, CBAM also has core environmental goals, aiming to encourage more countries around the world to adopt carbon pricing measures, and therefore to incentivize decarbonization.
While the positive impacts of CBAM have already begun, with several countries announcing domestic carbon pricing schemes in response to the policy, the EU has also faced significant backlash from its trading partners. Specifically, CBAM has been criticized for disproportionately affecting low- and middle-income countries, which often depend on carbon-intensive exports and may lack the financial resources to decarbonize in response to CBAM. In addition, the lack of domestic carbon pricing systems in most developing nations has left them unable to safeguard CBAM revenues through their own effective carbon price (ECPs), which firms can deduct from their CBAM dues. Critics therefore argue that CBAM risks exacerbating global inequalities, conflicting with principles such as Common but Differentiated Responsibilities.
To address these criticisms, this paper proposes expanding CBAM’s definition of an effective carbon price to account for investment in decarbonization. We suggest this could be enacted by recognizing the retirement of certified “carbon tax assets” representing emissions reductions from carbon mitigation projects such as clean energy. These “carbon tax assets” could be used by firms to match “carbon tax liabilities” arising from CBAM. Exporters to the EU could therefore reduce their CBAM bills by purchasing carbon tax assets, generating investment for decarbonization projects in the Global South.
It is crucial to note that our proposal preserves CBAM’s core objective of maintaining a competitive level playing field by denominating carbon asset deductions in monetary terms, rather than directly offsetting tons of carbon emissions. By linking CBAM deductions to the price paid for carbon assets, exporters face equivalent economic pressures to reduce emissions, avoiding risks of carbon leakage and perverse incentives towards low-quality assets. In other words, unlike previous links between carbon assets and carbon tax liabilities, for example the use of Clean Development Mechanism (CDM) credits in the EU ETS, our proposal does not simply allow a ton to offset a ton, but rather recognizes that the price paid for an offset can be considered a tax, and thus deducted as an ECP.
Mechanically, our proposal operates through two channels to support developing countries burdened by CBAM. First, the “external channel”, where exporters in developed nations purchase carbon assets from developing countries to pay their CBAM liabilities. This might be, for instance, a Japanese steel company exporting to the EU investing in a clean energy project in Mozambique. In this way, the external channel creates a new flow of external carbon finance from Global North to Global South by generating demand for carbon assets.
Second, we propose an “internal channel,” where developing nations can use our flexible definition of an ECP to retain carbon revenues and encourage green investment at home. This would allow domestic firms to offset CBAM obligations through investments in local decarbonization initiatives, for example a steel producer in Azerbaijan funding clean manufacturing and deducting its investment from its CBAM dues.
This proposal comes at a critical time both for border carbon adjustments and for climate finance. At the international level, the recent agreement on Article 6.4 methodologies at COP29 could provide a starting point to define robust eligibility standards for carbon assets. At the same time, the inclusion of border carbon adjustments as part of the COP29 agenda for the first time, within a Presidency’s Consultation to report by COP30, offers a multilateral forum to discuss how this proposal might be enacted.
For the EU, while the Article 6 methodologies establish a foundational framework for international carbon markets, there is the opportunity to “gold-plate” international standards with additional criteria on which carbon assets would count as an ECP. This gold-plating could include metrics such as additionality, permanence, transparency, social impact, and governance. This could enhance carbon market integrity and reinforce the EU’s leadership in global climate governance.
To operationalize our proposal, this white paper contains an appraisal of the next steps that will be necessary from both the EU and its trading partners. For the EU, these include issuing delegated or implementing acts under the CBAM regulation to explicitly recognize carbon mitigation assets as ECPs. As we discuss, it is also possible to directly amend the CBAM regulation via changes to Articles 2 and 9. Simultaneously, EU trading partners will need to enact national legislation governing the use of carbon assets, ensuring alignment with EU criteria. This could involve establishing registries for carbon asset certification and leveraging international standards like those defined at COP29.
Border carbon adjustments, starting with the EU CBAM, represent a huge opportunity for the energy transition, creating incentives for decarbonization and revenue for governments. But for these policies to reach their potential, equity concerns amongst developing country exporters must be addressed. Our hope is to link two great challenges for the energy transition – the need for carbon pricing and the need for climate finance – letting one create a solution for the other. Allowing carbon assets to match carbon liabilities could turn CBAM into both a carrot and a stick. This aims to present a practical and just way forward.Version of Recor
Transitional Justice after BlackLivesMatter: A Comparative Case Study of Truth Commissions for Racial Justice.
No full textIn the summer 2020, “BlackLivesMatter” demonstrations led to the largest civil unrest across the U.S. since the civil rights movement. Campaigners protested against the disproportionate number of killings and arrests of black Americans by the police. These protests brought to light the issue of institutional racism, i.e., hidden forms of racism that permeate American society and are embedded in law and policy.
Truth commissions are increasingly used to deal with issues of racial justice. Such commissions are a mechanism of transitional justice, a holistic and restorative approach to justice whereby societies seek to come to terms with a legacy of serious human rights abuses. Given the petitioning for establishing a truth commission for racial justice after the “BlackLivesMatter” protests, this thesis investigates whether truth commissions provide the adequate tools to address institutional racism. Its main objective is to investigate whether truth commissions improve legislative reforms in the U.S.
This thesis therefore seeks to answer the following research question: do truth commissions have a measurable impact on government policy and legislative change? Very few studies have attempted to measure the impacts of truth commissions on legal reforms. Accordingly, this thesis seeks to identify and explain the causal mechanisms through which truth commissions (supposedly) drive legislative reforms.
To do so, it undertakes a comparative case study of selected truth commissions tasked with issues of racial justice and process-traces their operations. This thesis employs a “root cause” theoretical framework. It finds that truth commissions, which conduct more sophisticated analyses of the root causes of racialized inequalities, understood as investigations into how laws, policies and practices leading to the unfair treatment of people of color have developed and persisted, exert direct influence on policymakers and generate better legal reforms. By contrast, this thesis argues that truth commissions, which solely focus on the individual allocation of guilt, rarely generate legal reforms
QuTiE: Universal, Declarative, and Differentiable Physics Simulation
No full textAs physical experiments and theories increasingly venture into regimes that are difficult to construct regularly, but nevertheless essential for scientific progress, a need is identified for a physics-focused domain-specific language (DSL) which can fully specify general physical systems, and translate the specification into efficient machine code to simulate the results of otherwise cost-prohibitive experimental setups. A novel, declarative language “QuTiE” for describing physical systems is presented. It is shown to be universal, by being able to simulate all quantum physics on an n-dimensional stratified manifolds, including quantum field theory and General Relativity. A reference implementation is also provided, which is able to simulate all quantum mechanical systems that previous simulators can, with the addition of automatically resizing spaces so that boundaries do not need to be known a priori and transparent, automatic multithreading which needs no intervention from the physicist.Computer Scienc
Robust and Multimodal Signals for Language in the Brain
No full textAn impressive aspect of human language is its ability to maintain consistent meaning and grammatical correctness across different sensory modalities. Elucidating the modality-invariant internal representation of language in the brain has major implications for cognitive science, brain disorders, and artificial intelligence. A pillar of linguistic studies is the notion that words have defined functions, often referred to as parts of speech, such as nouns and adjectives. This dissertation tries to answer two questions: how can we find a multimodal and generalizable representation for language processing? Is there a representation for parts of speech in the brain?
To address these questions, I recorded neural responses from 1,801 electrodes in 20 participants with epilepsy while they were presented with two-word minimal phrases consisting of an adjective and a noun, in both auditory and visual presentations. I observed neural signals that distinguished between these two parts of speech (POS), localized within a small region in the left lateral orbitofrontal cortex. The representation of POS showed invariance across several criteria: visual and auditory presentation modalities, and robustness to word properties like length, order, frequency, and semantics. The results also generalized across two different languages in a bilingual participant. I found that these selective signals provide key elements for the compositional processes of language, highlighting a localized and invariant representation of POS. Furthermore, I extended these ideas by evaluating how parts of speech are processed within full sentences. Recording activity from 1,563 electrodes in 17 participants, I found neural signals that separated nouns from verbs in sentences. This selective, invariant, and localized representation of parts of speech provides key elements for the representation of language. Going beyond POS, I also found multimodal representations for grammar and syntax processing in sentences, a first with neurophysiological signals.
In conclusion, I collected and analyzed a dataset of over 60 hours of stimuli, 40,000 two-word phrases, and 15,000 sentences by recording neural responses to audiovisual stimuli from 3,394 stereo-electroencephalography electrode contacts across 37 participants. My findings report neural representations of language that are both robust and adaptable, contributing to a deeper understanding of core linguistic processes within the human brain. This work lays a foundation for more nuanced studies of language processing. To the best of my knowledge, this is the first study to rigorously and systematically evaluate the invariant representation of parts of speech and multimodal representations for semantic and grammatical processing of sentences at the invasive neurophysiological level from the human brain.Biophysic