Photo: Robbie Sproule from Montreal, Canada / Wikimedia Commons / CC BY 2.0
Boolean operators refine legal database searches effectively. They let a researcher control how search terms work together. The main operators are AND, OR, and NOT. Legal databases may also offer proximity connectors, which limit the distance between words.
This matters because legal language shifts. A court may use “lawyer,” “attorney,” or “counsel.” A case may discuss both “privacy” and “personal data,” but the words may appear far apart. A plain keyword search can miss useful records or return too many weak ones.
I think of a search as a claim about language. Each operator states what must happen in the result.
AND requires both terms. A search for privacy AND workplace looks for records containing both ideas. This can narrow a broad set of results.
OR joins related terms. A search for lawyer OR attorney OR counsel allows any of those words. This often helps with synonyms and older language. It also makes a search wider.
NOT removes a term. A search such as contract AND NOT employment may remove records that discuss employment contracts. That can help, but exclusion needs care. A useful case may mention the excluded word while still addressing the main issue.
The order of terms also matters in some systems. Parentheses can group ideas, as in (lawyer OR attorney) AND (privacy OR surveillance). The search then asks for one term from each group. Without clear grouping, a database may read the query in a different order than the researcher expects.
Legal databases often add proximity commands. These require terms to appear near each other. Lexis documentation describes connectors such as W/5, which finds words within five words of each other, and PRE/3, which requires one term to come before another within three words. Westlaw documentation describes similar paragraph, sentence, and word-distance connectors.
A proximity search can be more precise than AND alone. Consider notice W/5 termination. It seeks records where “notice” and “termination” occur close together. A search for notice AND termination may return a document where the terms appear in separate sections.
This difference is easy to miss. AND tests whether terms appear somewhere in the same record. A proximity connector tests their local relationship. The second question is often closer to the legal issue.
The search terms still carry the main burden. Operators cannot repair poor terms. A search for wrongdoing AND law may be too broad. A search for (whistleblower OR "qui tam") AND retaliation states the subject more clearly, assuming those terms fit the research question.
Phrases can help as well. Quotation marks often search words together as a set, though the exact rule depends on the database. Some systems treat an unmarked group of words as a phrase. Others may interpret it through a different search mode. The search guide for the database controls this point.
That is where caution enters. Legal research platforms do not all use the same syntax. Lexis materials describe both natural-language searching and a terms-and-connectors method. Westlaw materials distinguish among connectors, proximity commands, and search forms. A symbol that excludes a term in one system may have another meaning elsewhere.
The database’s coverage creates a second limit. A precise query can still miss relevant law if the database lacks a needed court, publication, date range, document type, or full text. Search skill cannot add records that the collection does not hold. A result list also does not prove that a search was complete.
This limit changes how Boolean searching should be judged. The goal is not a single perfect query. The goal is a set of clear searches that test different wording choices. One search may use narrow terms. Another may add synonyms. A third may use proximity. The pattern of results can show where language changes.
The results also need human review. A record may contain every required term and still have little value. It may mention the issue only in a citation, background section, or opposing argument. Boolean logic controls retrieval. It does not decide legal relevance.
I would record the exact query, the database name, and the search date. Database interfaces and rules can change. A saved search may not behave the same way later, especially when a publisher changes its search system or updates its help pages.
The plain lesson is simple. AND narrows. OR gathers language variants. NOT excludes. Proximity connectors test whether words appear near each other. Together, these tools turn a broad legal database search into a stated method.
The honest limit is just as simple. Boolean operators improve control, but they do not guarantee complete research. Coverage, syntax, indexing, document language, and human judgment still shape the result. For that reason, a strong search is precise without pretending to be final.
That balance fits The Source List: one digital source worth knowing, one search tip, and one honest limitation. A useful legal research tool is easier to trust when its search rules and boundaries are stated plainly.