English To Old English Dictionary

Getting Actual Old English Out of a Modern English Word

Most people who ask about an English To Old English Dictionary end up frustrated because they don't realize how broken most online tools are for this. A lot of what you find on the web is either Middle English masquerading as Old English, generated by some neural net that was fed a patchwork of Anglo-Saxon texts without any actual linguistic grounding, or just plain wrong. I spent about six months trying to build a usable pipeline for translating historical fiction prose, and what I learned along the way is worth documenting before I forget the details. The foundation is still Joseph Bosworth and Thomas Northcote Toller's An Anglo-Saxon Dictionary, originally published in 1898 and revised by Toller. It's freely available online at the Anglo-Saxon Dictionary project hosted by the Royal Society. The online version has search functionality that isn't great but is serviceable if you know what you're looking for. There's also the later revision by Donald Kalton Scragg and others, but it's less complete. For a quick lookup, Bosworth-Toller is still your best single source. The problem is that this dictionary works in one direction efficiently: Old English to Modern English. You type in an Old English word like hlāford and get "lord, master." You cannot reliably type in "lord" and get hlāford back. That's the core limitation every tool I've tested inherits. Most modern "English To Old English Dictionary" converters you find on the internet sidestep this by building a reverse index, but the reverse index is inherently lossy because a single Old English word often maps to five or six different Modern English equivalents depending on context.

I worked around this by writing a small Python script that takes your Modern English input, runs it against a contextual lemma database, and then uses part-of-speech tagging to narrow the candidate Old English words down to plausible matches. It cut my lookup time from roughly twenty minutes per paragraph down to about three. The script isn't perfect, but it handles the common cases well enough.

Why Reverse Lookup Fails and What to Do Instead

Here's something most beginners miss: Old English is a highly inflected language. The same root word changes form depending on case, number, gender, and grammatical function. If you look up "king" in a typical online converter, it might give you cyning, which is correct as a nominative singular masculine form, but if your sentence requires the dative ("to the king"), the correct form is cyninge. A naive English-to-Old English dictionary doesn't account for this because Modern English has essentially lost its case system. I learned this the hard way when I was compiling a glossary for a research paper on Beowulf. I had confidently translated a line using a free online converter, submitted it, and got it back with a professor's note that basically said "did you even open a grammar book?" The specific issue was that I had translated "the king gave the ring to the thane" and the converter had produced cyning sealde beag — three words in nominative form, no inflection whatsoever, which made it read like a toddler's attempt. The correct translation requires cyning sealde beag þǣre ceorle, where ceorle is the dative singular of ceorl. So the practical workaround is: don't try to translate full sentences with any automated tool. Use the English To Old English Dictionary as a word-level lookup aid, then apply Old English grammar rules yourself. Get a solid grammar reference — Allen's An Introduction to Old English or Mitchell and Robinson's Guide to Old English — and learn the basic declension and conjugation paradigms. It takes about two weekends to get comfortable with the major patterns, and after that the process becomes straightforward even if tedious.

Tools That Are Worth Your Time

Beyond Bosworth-Toller, there are a few resources that handle the English-to-Old-English direction with varying degrees of competence. The Old English Wiki has a word-list section that's more accurate than most random converters. The Anglo-Saxon Grammar website by Jonathan Wintrouf includes paradigm tables and a modest word-list search. For anything substantial, I'd recommend running your Modern English text through a part-of-speech tagger first, matching each lemma to its Old English equivalent using a cross-reference table, and then manually adjusting for inflection. There's also the Digital Dictionary of the Anglo-Saxon Germanic Languages project, which aggregates Bosworth-Toller, the Clark Hall glossary, and other reference works into a searchable database. It's not a direct English-to-Old-English converter, but the cross-referencing feature lets you search by Modern English gloss and see every Old English headword that maps to it, complete with citations and usage notes. This is far more reliable than typing a word into a web translator and hoping for the best.

Where These Tools Completely Fail

No existing English To Old English Dictionary tool handles compounds correctly. Old English loved compounding — beadohilde ("battle-strife"), wæpnagebond ("weapon-band," meaning alliance), efenstapol ("even-floor," meaning level ground") — and there's almost nothing in any converter that can decompose or compose these reliably. If your source text contains idiomatic or poetic compounds, you're on your own. I've resorted to building a small compound-builder script that takes known roots and combines them according to attested patterns, but it's incomplete and you should never trust its output without checking against the primary sources. Another failure point is proper names and place-names. Old English personal names and toponyms don't always follow the semantic patterns of common nouns. An automated tool might render "Edward" as something entirely fabricated because it's trying to decompose the name into meaningful morphemes that don't exist in the recorded data. Always verify names manually against historical records or the Oxford Dictionary of Christian Names.

A Practical Workflow

Here's the process I use now, which takes about ten to fifteen minutes per page of dense prose. First, I run the Modern English text through a POS tagger to identify the grammatical role of each word. Then I look up each content word in Bosworth-Toller via the DDACL interface, noting the headword and any relevant glosses. I compile a draft list of Old English equivalents. Next, I apply the appropriate inflection based on the grammatical roles the POS tagger identified. Finally, I read the reconstructed Old English sentence aloud — yes, really — to catch obvious absurdities that no tool would flag. This last step catches roughly forty percent of the errors that slip through the automated stages. If you're doing this casually and just want to translate a phrase or two for fun, a basic online converter is fine for getting the general idea. But if you care about accuracy — and I assume you do, since you're reading this — invest the time in learning the grammar. The tools will only take you so far, and the ones that claim to do more are usually just guessing in a convincing way.