Source Use & Research Strategies for Extemporaneous Speaking

Establish credibility with judges: demonstrates you are informed and trustworthy.
Serve as external verification so you avoid spreading misinformation.
Uphold the educational purpose of extemp: to inform accurately.
Competitive necessity: many tournaments now run live or post-round fact-checks (pre-lims, semis, finals; e.g., NSDA Nationals, Blue Key, Round Robin).
• Ballots often include a “source-check passed/failed” line.
• A failed check can drop you to a $6^{th}$ rank regardless of speaking quality.
Moral dimension: refusing to lie counters broader societal disinformation (e.g., politicians denying climate change).

Nationals: a designated judge writes down every publication and date you cite.
Other contests increasingly replicate the practice; expect random audits.
Judges may place a “?” beside anything that looks off and Google it later.
Consequence of fabrication or wrong dates: automatic low ranks, possible tournament penalties, damaged reputation (“children making stuff up”).

Reading and connecting information physically builds brain structures; ChatGPT cannot build neurons for you.
Active processing → stronger recall when you are cross-examined.

Ideal total per speech: $7$ .
• $1$ in the background section of the introduction (frames the question and shows you actually researched).
• $2$ per body point (Point I, II, III) → $2 \times 3 = 6$ .
Beginners may start with fewer, but build to seven; advanced competitors may add a third citation per point (≈ $10$ total) once memory skills are solid.
Full verbal citation goal: Source, Month, Day, Year (e.g., “According to the Council on Foreign Relations, $03/17/2025$ …”).
• New extempers may begin with Month + Year, but advance toward the complete date.
• Some judges explicitly dock for missing the day.

Never mash two sources together (“according to the IMF and Brookings…”).
• You lose the persuasive power of showing independent confirmation.
Blogs are risky; only use if it is an official institutional blog (e.g., “Johns Hopkins Medicine News & Views”).
If citing unfamiliar think-tank acronyms, briefly clarify: “BICSET, a Bangladeshi green-tech think tank…”.
Congressional hearings: preferably quote a mainstream report summarizing the hearing; otherwise cite the committee name, subject, and full date.
Tweets or social media: permissible if you identify the platform and user (“Brazilian President Jair Bolsonaro tweeted $04/12/2025$ …”), and only when primary-source value is essential.

Think tanks dedicated to informing policymakers (CFR, Brookings, Cato, CSIS, RAND). Check ideological tilt first.
Established newspapers with strong editorial standards (Wall Street Journal, Washington Post, New York Times, Guardian). Often require subscriptions → avoid paywalls in prep by arranging school/team access.
Newswires: Associated Press, Reuters (pronounced “ROY-ters”).
Publicly funded broadcasters: NPR, BBC, PBS.
Regionally relevant outlets: Austin American-Statesman for Texas floods, Al Jazeera for Middle-East, local African papers for IX Africa question. Judges appreciate geographic specificity.
Primary sources: official State Department releases, central-bank statements, agency white papers.

Centered / factual → safest.
Far-left or far-right commentary (Daily Beast, Breitbart, Young Turks, Tucker Carlson) may work as ATD examples but not as core evidence.
Always ask: would an unknown judge immediately accept this outlet as legitimate?

Prefer articles ≤ 1 year old.
Exceptions: historical contrast (“10 years ago…”) or Past–Present–Future analytical structure—but still add fresh data.
Final-round speakers often flaunt hyper-currency: “Thirty minutes ago Bolsonaro tweeted…”.

Pair ideologically diverse outlets to show consensus (e.g., Cato Institute + New York Times both criticizing a Trump policy).
Mixing domestic and regional papers demonstrates depth.
Avoid repeating the exact same publication too many times (“NYT, NYT, NYT…”). If you must, use clearly different reporters and dates.

Extemp Genie
• $≈\$25$ subscription, autofiles into categories, works offline (lifesaver when tournament Wi-Fi fails—e.g., Nationals Octafinals fiasco).
Card-cutting (old-school)
• Printing, highlighting, and sorting articles into physical tubs.
• Guarantees offline access & deep familiarity but time-consuming and tree-heavy.

Read daily. Your phone grants omnipresent access—use idle moments.
Podcasts for audio learners: NPR “Up First” (~ $14$ min), BBC “Global News”, specialized shows (e.g., “The Diplomat Asia Pacific”).
Email headline digests (NYT, Axios, Foreign Policy) to maintain baseline awareness even on busy school days.
After reading, consciously connect dots: patterns, cause–effect, future scenarios.
Study history (e.g., colonialism) to avoid ignorant claims and to enrich modern analysis.

Pre-tournament: read background so the question is never your first exposure.
In-prep tips (from recent circuit grads):
• Search broad term + sub-angle ("Pakistan infrastructure agriculture 2025").
• Open 2–3 promising articles simultaneously, look for overlap to craft sub-points.
• Use site:gov, site:edu to filter credible domains.
• If paywalls appear, pivot rapidly; 30-minute prep time is finite.

Generative AI can brainstorm outlines but:
• Provides “glop” that you didn’t critically evaluate; neural pathways not formed.
• Hallucinates facts and fake citations—fatal under cross-ex or source check.
• Cannot feed you real-time answers while you speak.
Use at your own peril; still verify every datum manually.

Polarizing positions (e.g., anti-Trump 2017 Harvard final) sometimes succeed, sometimes tank.
Know the risk: judge ideology is unknown.
Decide whether you’re in the room to win or to say what you believe; opportunity cost is personal.

$7$ total citations (minimum guideline).
$2$ per body point.
Intro must have $1$ citation (background).
Podcast “Up First” ≈ $14$ minutes.
30-minute prep window → allocate time between research and rehearsal (don’t chase the “perfect source” so long you never practice).