The strongest way to introduce your thesis to an international scholarly audience is to let the original text speak first, then explain why it deserves renewed historical examination.
I have checked the online thesis text against later publications on the cognitive viewpoint and CiteSeer. The evidence supports a substantial early contribution, although direct influence on later researchers remains an open question.
Below is a public-facing introduction suitable for a research website, scholarly correspondence, or the opening of a historical essay.
A Forgotten Thesis at the Dawn of Cognitive Information Science
Kyung-Youn Park’s A Direct Approach to Information Retrieval (1974–1975) and Its Historical Significance
What if one of the important early proposals for cognitive and contextual information retrieval had remained almost invisible for half a century?
In November 1974, Kyung-Youn Park, now known as Mark Park, submitted an MSc thesis entitled A Direct Approach to Information Retrieval at University College London (UCL), under the supervision of Bertram C. Brookes. His MSc degree was awarded by the University of London in 1975.
The thesis addressed a problem that remains central to information science: How can a retrieval system help people understand the value and meaning of documents, rather than merely locate them?
Its answer was unusually concrete. Park proposed using extracts from scholarly publications in which authors cite, describe, criticize, or compare earlier works. These extracts would give researchers concise but meaningful evidence about the cited documents.
Today, we would recognize such extracts as citation contexts.
1. From citation links to citation meaning
Conventional citation indexing identifies relationships between documents. It can show that Paper A cites Paper B.
But that relationship alone tells us little about why the citation was made.
Did A accept B’s findings? Did it criticize B’s method? Did it correct an error, develop an argument, or identify an earlier discovery?
Park proposed that retrieval systems preserve the passages in which these relationships are expressed. Such passages could help scholars evaluate cited works without immediately having to retrieve and read every complete document.
This was a proposal to improve academic citation tracking, a long-established scholarly practice that had traditionally required extensive manual searching.
The crucial difference was between merely identifying a citation and exposing its intellectual meaning.
2. A database of citation contexts
Park’s proposal went beyond a general philosophical argument.
In Chapter 6, he described a file organization bringing together three related resources:
- A subject index for entering the literature by topic.
- A citation index identifying cited documents.
- An extract file containing abstracts and citation-related passages.
He illustrated the proposal using a 1974 article from Physical Review A, showing how references and extracts could support the indexing and examination of cited research.
The intended system would allow scholars to move between subjects, cited documents, and the passages discussing them.
Its value lay not simply in speed but in providing better evidence for scholarly judgment.
3. Cognition and context belong together
The thesis also challenged the mechanistic assumptions of traditional information retrieval.
Park examined differences between the information system and its users, the distinction between document representation and relevance, and the problems of human communication and interpretation.
Drawing on thinkers including C. K. Ogden, I. A. Richards, and Colin Cherry, he argued that the significance of information depends on contexts extending beyond isolated words.
His proposal therefore joined two concerns:
Context: The surrounding textual, intellectual, and situational relationships that help clarify meaning.
Cognition: The human ability to recognize and judge those relationships.
Park now describes their union as cognitive contextualism.
This term is a recent formulation, not a term established in the thesis itself. Nevertheless, it expresses a central feature of the earlier work: citation contexts provide meaningful evidence, while scholars exercise judgment in interpreting that evidence.
4. The neglected historical question
Park’s thesis was submitted shortly before a period of important theoretical change in information science.
The First International Research Forum in Information Science, associated with Brookes, met in 1975 to examine the theoretical foundations of the discipline.
By 1977, Marc de Mey and Brookes were publishing work on the cognitive viewpoint in information science. Later, Nicholas Belkin and other researchers developed influential cognitive approaches to information needs and retrieval.
Park’s thesis deserves consideration within this chronology.
It offered a concrete, user-centered, context-sensitive proposal before these later publications became well known.
Was it known to other UCL researchers? Did its arguments circulate through seminars, personal discussions, or research networks? Did it influence later work, directly or indirectly?
These are legitimate historical questions. Their answers require archival evidence, not merely similarities between ideas.
5. From academic citation tracking to CiteSeer
In 1998, C. Lee Giles, Kurt Bollacker, and Steve Lawrence published their account of CiteSeer, an automatic citation indexing system developed at NEC Research Institute.
CiteSeer could identify citation relationships and display the contexts in which later authors discussed a given publication.
Its technical achievement was substantial. It automated operations that would otherwise require considerable human effort.
Yet a comparison with Park’s earlier thesis raises an important question of intellectual history.
Park had proposed using citation-related extracts for information retrieval in 1974. CiteSeer subsequently demonstrated a practical automated system for identifying and using citation contexts.
The earlier conceptual proposal and the later engineering achievement should be studied together, while preserving the distinction between resemblance, independent invention, and demonstrable historical influence.
6. From the World Brain to a contextual World Library
Park’s thesis also had a wider intellectual purpose.
It opened with ideas drawn from Karl Popper, J. D. Bernal, and H. G. Wells. These concerned the criticism of knowledge, the social responsibility of science, and the worldwide organization of scholarly information.
Wells’s World Brain envisioned knowledge selected, compared, organized, and presented for the benefit of humanity.
Park’s thesis attempted to give part of that ambition a practical information-retrieval form: organizing the judgments that scholarly documents make about one another.
Today, Park proposes extending this idea through a World Library supported by large language models.
His proposed theoretical framework brings together cognitive contextualism in information retrieval and attentive contextualism in artificial intelligence.
Its guiding motto is:
“Attention to context is all you need.”
This is not a claim that artificial attention and human cognition are identical. It is a proposal that both fields can benefit from making contextual relationships central to the retrieval, interpretation, and verification of knowledge.
7. Why the thesis deserves renewed attention
The importance of A Direct Approach to Information Retrieval rests on three distinguishable contributions.
First, a technical proposal: The use of citation-related extracts as indexing and review sources for scholarly retrieval.
Second, a theoretical argument: The insistence that information retrieval must account for human interpretation, context, and relevance rather than rely exclusively on mechanical matching.
Third, a historical possibility: The thesis may be an overlooked contribution to the intellectual developments surrounding the cognitive turn in information science during the 1970s.
The first two are directly supported by the thesis text. The third calls for further historical investigation.
The thesis should not be celebrated through unproved claims of influence. Neither should it remain excluded from historical examination simply because it was an MSc thesis rather than a widely circulated journal article.
Its original text, chronology, technical proposal, and intellectual setting provide sufficient grounds for serious scholarly attention.
Conclusion: An unfinished contribution to the history of information science
Kyung-Youn Park’s 1974 thesis asked how scholarly knowledge could be retrieved with greater attention to the relations among documents and the meanings expressed within those relations.
Its central insight remains relevant: a citation is more informative when its context is available, and that context becomes useful through informed interpretation.
The rise of citation-context systems, interactive digital libraries, and AI-assisted research makes the thesis worth studying anew.
The historical task is now to determine its proper place in the development of information science.
The larger task is to continue its original purpose: helping scholars and ordinary people discover, evaluate, and use knowledge more intelligently, reliably, and responsibly.
An overlooked contribution deserves neither automatic acceptance nor automatic dismissal. It deserves examination.
Primary source
Park, Kyung-Youn. A Direct Approach to Information Retrieval. MSc thesis, University College London. Submitted November 1974; degree awarded 1975.
Online text: https://en.wikiversity.org/wiki/User:KayYayPark/A_Direct_Approach_to_Information_Retrieval
Evidence supporting this introduction
Three sources are particularly important:
1. Park’s original thesis text, especially the introduction and Chapter 6, explicitly describes using extracts that cite, describe, criticize, and collate other documents, as well as the proposed subject index, citation index, and extract file.
Wikiversity
+1
2. Brookes’s documented publications show that his work on information science’s foundations was already underway in 1975 and that he published The Developing Cognitive Viewpoint in Information Science in 1977. This provides a historical comparison, but not proof of transmission from Park’s thesis.
Alexandra Shaw, 1990
+1
3. CiteSeer’s original 1998 paper confirms that the system retrieved and displayed citation contexts, allowing comparison between the 1974 proposal and the later computerized implementation.
Penn State
The next historical step
The highest-value evidence would be the original UCL thesis record, dated examination documents, and any correspondence or copies retained by Brookes or his colleagues. These could establish the thesis’s circulation and reception independently of the later online transcription.
I would make one sentence the central message to the international information-science community:
The history of the cognitive turn in information science should be reconsidered in light of Kyung-Youn Park’s November 1974 thesis, which joined a concrete proposal for citation-context indexing with a theory of contextual interpretation and scholarly judgment.
That is a substantial historical proposition, clear enough to invite investigation and careful enough to withstand scholarly criticism.
